Skip to main content

Full text of "Analyzing environmental and structural characteristics of concrete for carbon mitigation and climate adaptation in urban areas: a case study in Rajkot, India"

See other formats


ANALYZING ENVIRONMENTAL AND STRUCTURAL CHARACTERSITICS OF 

CONCRETE FOR CARBON MITIGATION AND CLIMATE ADAPTATION IN 

URBAN AREAS: A CASE STUDY IN RAJKOT, INDIA 

by 

Andrea Valdez Solis 

B.S., New Mexico State University, 2006 

M.S. New Mexico State University, 2008 



A dissertation submitted to the 

Faculty of the Graduate School of the 

University of Colorado in partial fulfillment 

of the requirements for the degree of 

Doctor of Philosophy 

Civil Engineering 

2013 



©2013 
ANDREA VALDEZ SOLIS 
ALL RIGHTS RESERVED 



This dissertation for the Doctor of Philosophy degree by 

Andrea Valdez Solis 

has been approved for the 

Civil Engineering Program 

by 



Stephan A. Durham, Chair 

Anu Ramaswami, Co-Advisor 

Arunprakash Karunanithi 

Ross Corotis 

Yunping Xi 



December 17, 2012 



11 



Solis, Andrea, Valdez (Ph.D., Civil Engineering) 

Analyzing Environmental and Structural Characteristics of Concrete for Carbon 
Mitigation and Climate Adaptation in Urban Areas: A Case Study in Rajkot, India 

Dissertation directed by Associate Professor Stephan A. Durham 



ABSTRACT 

Increasing temperatures, varying rain events accompanied with flooding or 
droughts coupled with increasing water demands, and decreasing air quality are just some 
examples of stresses that urban systems face with the onset of climate change and rapid 
urbanization. Literature suggests that greenhouse gases are a leading cause of climate 
change and are of a result of anthropogenic activities such as infrastructure development. 
Infrastructure development is heavily dependent on the production of concrete. Yet, 
concrete can contribute up to 7% of total CO2 emissions globally from cement 
manufacturing alone. 

The goal of this dissertation was to evaluate current concrete technologies that 
could contribute to carbon mitigation and climate adaptation in cities. The objectives 
used to reach the goal of the study included (1) applying a material flow and life cycle 
analysis (MFA-LCA) to determine the environmental impacts of pervious and high 
volume fly ash (HVFA) concrete compared to ordinary portland cement (OPC) concrete 
in a developing country; (2) performing a comparative assessment of pervious concrete 
mixture designs for structural and environmental benefits across the U.S. and India; and 
(3) Determining structural and durability benefits from HVFA concrete mixtures when 
subjected to extreme hot weather conditions (a likely element of climate change). 



111 



The study revealed that cities have a choice in reducing emissions, improving 
stormwater issues, and developing infrastructure that can sustain higher temperatures. 
Pervious and HVFA concrete mixtures reduce emissions by 21% and 47%, respectively, 
compared to OPC mixtures. A pervious concrete demonstration in Rajkot, India showed 
improvements in water quality (i.e. lower levels of nitrogen by as much as 68% from 
initial readings), and a reduction in material costs by 25% . HVFA and OPC concrete 
mixtures maintained compressive strengths above a design strength of 27.6 MPa (4000 
psi), achieved low to moderate permeability's (1000 to 4000 coulombs), and prevented 
changes in length that could be detrimental to the performance of the concrete in long- 
term temperatures above 37.8°C (100°F). 

The form and content of this abstract are approved. I recommend its publication. 

Approved: Stephan A. Durham 



IV 



DEDICATION 

I dedicate this work to my parents Loretta Valdez and Andrew Chavez and to all 
the people from the pueblitos of Northern New Mexico. The love, care, and support 
these people show help others strive for the best, believe, and remain positive in life. 



ACKNOWLEDGEMENTS 

I would like to thank my advisors Dr. Stephan Durham and Dr. Anu Ramaswami. 
My advisors provided me a unique PhD experience that has taught me how to be a 
stronger person both in life and in my profession. The PhD was challenging but, Dr. 
Duhram and Dr. ramswami helped me to realize the importance of remaining patient, 
motivated, and grateful while doing research. I am honored to have studied under the 
guidance of these two very important people who are admired for their personalities and 
contributions to engineering and sustainability. I come away with a PhD striving to 
model the best attributes of my advisors, Dr. Durham for his practicality, passion for 
teaching, and appreciation he shows to others and Dr. Ramswami for her devotion and 
dedication she puts into every project, ability to challenge and motivate you with her 
words, and the courage they both display in being leaders in research. 

I would like to emphasize that the PhD experience was feasible and memorable 
because of the opportunity to meet and work with various people. If it wasn't for the 
times spent drinking tea, talking to and joking with fellow students and staff, or learning 
about cultures and collaborating with people across the world I would have overlooked 
how exceptional and distinct each person is in this world. It so important to learn how to 
work with different people and appreciate that chance to listen to their ideas, knowledge, 
concerns, and joys. I want to thank Tom Thuis, Randy Ray, Dr. Nien-Yin Chang, Dr. 
Kevin Rens, Dr. Rajaram, Jose Solis, Adam Kardos, Dr. Loren Cobb, Dr. Angie Hager, 
Derek Chan, Dr. Rui, Liu, Devon, Krista Nordback, Brian Volmer for all their help 
during my research and dissertation preparation. I thank Laasya Bhagavatula, Emani 
Kumar, Ashish Rao Ghorpade of ICLEI-South Asia, Mr. Jayant Lakhlani of Lakhlani 



vi 



Associates, Mitesh Joshi and his family and Alpana Mitra and her family for making me 
feel welcomed in Rajkot, India and giving me the honor of working with all of you while 
doing the research in Rajkot. Additionally, I appreciate the feedback and commitment 
that my committee members (Dr. Ross Corotis, Dr. Arunprakash Karunanithi, and Dr. 
Yunping Xi) showed during defense. I would also like to thank the National Science 
Foundation's Integrative Graduate Education and Research Traineeship (IGERT Award 
No. DGE-0654378) for funding my research. 

Lastly, I thank my family, friends, and especially my parents. It is hard to explain 
how much I appreciate the qualities of my parents because my parents mean a lot to me 
and I want to say the right words. My mom is always forgiving, a great listener, and I 
admire her for her ability to manage people and make people feel important. My dad is a 
very intelligent man that enjoys the simple things in life (like working side by side with 
his children), he gives valuable advice and I admire him for how hard he works. I am 
able to achieve any goal because my parents have always been there pushing me along, 
keeping me focused, and making me believe I have a purpose. 



vn 



TABLE OF CONTENTS 

Chapter 

1. Introduction 1 

1.1 Concrete and Urban Infrastructure 1 

1.1.1 Concrete Use 1 

1.1.2 Concrete Infrastructure Is a Source of GHG Emissions 3 

1.2 Climate Change in Urban Areas 3 

1.2.1 Flooding or Drought in Urban Areas 4 

1.2.2 Extreme Temperatures in Urban Areas 5 

1.3 Concrete Infrastructure for GHG Mitigation and Climate Adaptation 6 

1.3.1 Pervious Concrete Past and Contemporary Research 7 

1.3.2 High Volume Fly Ash Concrete Research with a Focus on Thermal Properties 9 

1.3.3 Main Goal and Knowledge Gaps 13 

1.4 Thesis Objectives 16 

1.5 Organization of Thesis 17 

2. Case Study Location: The City of Rajkot India 19 

2.1 Demographics, Population, and Climate 19 

2.2 Rajkot Construction and Concrete Infrastructure 21 

2.2.1 Personal Account of Construction 22 

2.2.2 Rajkot Concrete Infrastructure 25 

2.3 Future GHG Mitigation and Climate Adaptation Goals 29 

2.3.1 Stormwater/Rainwater Harvesting 31 

2.3.2 HVFA Concrete Road Project 32 



vin 



2.3.3 Collaboration between UC Denver, ICLEI South Asia 

and Rajkot Municipal Corporation 34 

3. Carbon Mitigation Through Concrete: An MFA-LCA Approach 36 

3.1 Bottom-line: Cement and Concrete Manufacturing in India and the US 36 

3.2 Life Cycle Assessment of Cement and Concrete in India 41 

3.3 Understanding the Cement Production and Concrete Industry in India 44 

3.3.1 Ready Mixed Concrete Industry in India 48 

3.3.2 Site Mixed Concrete in India 49 

3.3.3 Indian Concrete Mixture Designs 51 

3.4 Cement Manufacturing Process in India 52 

3.4.1 Phases of Cement Clinker 54 

3.4.2 Kilns 55 

3.5 Energy Consumption within the Cement Industry 56 

3.5.1 Energy Scenario in the Indian Cement Industry 57 

3.5.2 Methods of Energy Efficiency 58 

3.6 Management, Energy Efficiency Ventures, 

and Emission Trends for Indian Cement Companies 62 

3.6.1 Energy Efficiency and Embodied in Cement Manufacturing in India 63 

3.6.2 Emission Trends in Cement Manufacturing in India 65 

3.7 Materials, Fuels, and Emissions Associated with Cement and Concrete 70 

3.7.1 Cement 70 

3.7.1.1 Overall Result 76 

3.7.1.2 Company to Company Comparison 77 

3.7.1.3 Cementitious Materials 79 



IX 



3.1 A A Energy Intensity 80 

3.7.1.5 CO2 Emissions Factor Conclusion 81 

3.7.2 Quarrying and Mining of Other 

Raw Materials (Excluding Limestone) 82 

3.7.3 Coarse and Fine Aggregate Crushing 82 

3.7.4 Tranp sportation of Materials 83 

3.7.5 On-site Mixed Concrete 84 

3.7.6 Summary of Life Cycle Inventories 85 

3.8 MFA-LCA of Cement Use in Rajkot 87 

3.9 MFA-LCA for Concrete Mixtures in Rajkot 88 

3.10 Summary 90 

4. Stormwater Solution Demonstration with Pervious Concrete: 

Structural and Environmental Tests 91 

4.1 Study Design and Laboratory Phase I Testing 92 

4.1.1 Material Properties 95 

4.1.2 Mixture Design 97 

4.1.3 Test Methods 98 

4.1.4 Phase I Laboratory Results 105 

4.2 Providing Stormwater Management Solutions in Rajkot, India: 

A Pervious Concrete System Demonstration 112 

4.2.1 Introduction 112 

4.2.2 Materials and Methods 116 

4.2.3 Test Methods and Results 124 

4.3 Laboratory Phase II Testing (Cubes Versus Cylinders) 136 

4.3.1 Batching and Curing Phase II Laboratory Samples 138 

x 



4.3.2 Sample Shape Effects on the Compressive Strength of Pervious Concrete 141 

4.3.3 Comparing Compressive Strength Results 145 

4.3.4 Discussion of Standard Deviations and Population 149 

4.3.5 Summary of Percent Porosity 150 

4.3.6 Summary of Hydraulic Conductivity 151 

4.4 Summary 152 

5. High Volume Fly Ash Concrete for Hot Weather Conditions: 

Structural and Durability Tests 155 

5.1 Literature Regarding Fly Ash Use in India 155 

5.1.1 Properties of Fly Ash 155 

5.1.2 Fly Ash Consumption in India 157 

5.2 Literature on HVFA Concrete for Hot Weather Conditions 159 

5.3 Phase I study for HVFA in Hot Weather Conditions: India and U.S. Comparison of 
Fly Ash Properties (Fly Ash Used in Rajkot, Gujarat, India and Denver, Colorado, 
U.S.) 164 

5.4 Phase II: Properties of HVFA and OPC Concrete When 

Subjected to Hot Weather Conditions 172 

5.4.1 Aggregate Temperatures 172 

5.4.2 Verifying Temperatures of HVFA and OPC Concrete During Hydration 175 

5.5 Phase III study for HVFA in Hot Weather Conditions: Laboratory Testing of 
Structural and Durability Properties 180 

5.5.1 Compressive Strength 186 

5.5.2 Modulus of Elasticity 191 

5.5.3 Resistance to Rapid Chloride-Ion Penetration 191 

5.5.4 Length Change 198 

5.6 Applying a Multiple Linear Regression Model to Determine the Significance of 

xi 



Testing Variables on HVFA Concrete versus OPC Concrete When Subjected to Hot 
Weather Conditions 202 

5.6.1 Background on Multiple Linear Regression 202 

5.6.2 Application of the Multiple Linear Regression Models 203 

5.6.3 Revision of Multiple Linear Regression Analysis with Original Data 21 1 

5.7 Summary of Strength, Permeability, Length Change, and Multiple Linear 

Regression 215 

6. Conclusions and Recommendations 218 

6.1 Conclusions 218 

6.1.1 Carbon Mitigation: An MFA-LCA Approach 218 

6.1.2. Climate Adaptation: Pervious Concrete 219 

6.1.3 Climate Adaptation: HVFA Concrete 220 

6.2 Contributions 221 

6.3 Recommendations and Future Research 222 

6.3.1 MFA-LCA Recommendations 223 

6.3.2 Pervious Concrete Recommendations 223 

6.3.3 HVFA Concrete Recommendations 225 

6.4 Final Remarks Regarding Sustainability 235 

References 236 

Appendix 

A 247 

B 253 

C 255 

D 258 



xn 



LIST OF TABLES 

Table 

1.1. Summary of the Benefits of Fly Ash Concrete 10 

3.1 Comparison of Energy Use per Tonne of Cement Between the U.S. Cement Industry 
and India's Grasim Industries 37 

3.2 Summary of Energy Use and Emission Factors from Direct and Indirect CO2 
Emissions between India and the U.S 39 

3.3 World Cement Production 2010 46 

3.4 Indian Cement Industry Information 48 

3.5 Mixture Proportions for Typical Grades of Concrete (Based on Saturated 

Surface Dry Conditions) 51 

3.6 Average Energy Use Between India 

and U.S. Cement Industry for 2009-2010 58 

3.7 Examples of Non-Hazardous and Hazardous Alternative Fuels 62 

3.8 Example Differences in Calcining Emission Coefficients 69 

3.9 Fuel and Electricity Raw Data Gathered for Calculation of Cement Emission 
Factor 71 

3.10 Country Specific Emission Factors Used in Calculating a Cement Emission 
Factor 73 

3.11 Density Values for Certain Fuels Used in Indian Cement Manufacturing 73 

3.12 MFA-LCA Data for Purchased Electricity 74 

3.13 MFA-LCA Data for Company Generated Electricity from Coal 74 

3.14 MFA-LCA Data for Company Generated Electricity 

from LDO/Furnace Oil 74 

3.15 MFA-LCA Data for Company Generated Electricity from Natural Gas 75 

3.16 MFA-LCA Data for Thermal Energy from Coal 75 

3.17 MFA-LCA Data for Thermal Energy from Light Diesel 75 



xm 



3.18 MFA-LCA Data for Thermal Energy from Furnace Oil 75 

3.19 MFA-LCA Data for Thermal Energy from High Speed Diesel Oil 76 

3.20 Cement Production from Major Cement Manufacturing Companies that 

Deliver to Rajkot, India 76 

3.21 Energy Consumption from Major Cement Manufacturing Companies that 

Deliver to Rajkot, India 77 

3.22 Emissions from Major Cement Manufacturing Companies that Deliver to 

Rajkot, India 79 

3.23 Fly Ash Consumption by Major Cement Companies who Deliver to Rajkot, 

India 80 

3.24 Production and Emissions From Quarry and Mining 82 

3.25 Emission Factors for Aggregate Crushing 83 

3.26 Emission Factors and Average Distance Travelled 

for Cement Transportation 83 

3.27 Emission Factors and Average Distance Travelled 

for Transport of Aggregate 84 

3.28 Emission Factors and Average Distance Travelled 

for Transport of Fly Ash 84 

3.29 Specifications of Concrete Mixer 85 

3.30 Summary of Emission Factors Leading Up to Concrete Mixing 86 

3.31 Reiner's (2007) Emission Factor Calculations for Ready Mixed Concrete 86 

3.32 Information Regarding Rajkot Cement Use 

and Total Emissions per Year 88 

3.33 MFA Data for M35, Pervious and HVFA Concrete Mixtures 89 

3.34 LCA Data and Total Emissions Calculations from an MFA-LCA on Concrete 
Mixtures 89 

3.35 Cement Material Content and MFA-LCA Emissions for Certain Concrete 
Mixtures 90 



xiv 



4.1 Chemical Properties of Cement along with Standard Limits 96 

4.2 Physical Properties of Cement Along with Standard Limits 96 

4.3 Mixture Proportions for Phase I Laboratory Testing 98 

4.4 Porosity of Samples from Mixture 1 and Mixture 2 (Reported in Percent) 105 

4.5 Average Hydraulic Conductivity for Mixture 1 and 2 108 

4.6aMixutre 1 Compressive Strength Results 109 

4.6b Mixture 2 Compressive Strength Results 110 

4.7 Mixture Proportions for Rajkot 120 

4.8 Batch Quantities 120 

4.9 Specific Gravity Values Provided used in the 

Pervious Concrete Mixture Design 120 

4.10 Results of the Calculated Percentage Voids 125 

4.11 Hydraulic Conductivity of the Pervious Concrete and System 126 

4.12 Results of Compressive Strength of Pervious Concrete Samples 129 

4.13 Water Quality Analysis of the Water from a Bore Well and Stream 131 

4.14 Additional Results of Stream Water Quality Tests 134 

4.15 Mixture Proportions for Phase II Laboratory Testing 138 

4.16 Specific Gravities and Absorption Capacities in Phase II Testing 139 

4.17 Compressive Strength Results for M3 145 

4.18 Compressive Strength Results for M4 146 

4.19 Cylinder to Cube Strength Ratio 

Based on Average Compressive Strengths 147 

5.1 Example of Chemical Composition of Fly Ash from Different Countries 
(Malhotra & Mehta, 2008) 156 

5.2 Year 2005 Production and Utilization of Fly Ash in India 158 

xv 



5.3 Chemical Analysis for Various Fly Ash Sources between the U.S. and India 165 

5.4 Mixture Proportions for HVFA Concrete in Rajkot 166 

5.5 Compressive Strength Results for Rajkot HVFA Concrete Samples 167 

5.6 Mixture Proportions for HVFA Concrete in Denver 169 

5.7 Fresh Concrete Properties for the HVFA Concrete Batch in Denver 169 

5.8 Compressive Strength Results for U.S. HVFA Concrete Samples 170 

5.9 Average Cylinder to Cube Compressive Strength Ratios for U.S. and Indian 

HVFA Concrete Mixtures 171 

5.10 Mixture Proportioning for Mixture Designs in Phase Ha Testing of HVFA and 
OPC Concrete 176 

5.11 Mixture Proportioning for HVFA and OPC Concrete Mixture Designs in 
Extreme Hot Weather Condition Testing 181 

5.12 ASTM Standards Used for Fresh and Hardened Concrete Tests 181 

5.13 Material Temperatures Before Mixing (And During Mixing for the Heated 
Aggregate Mixtures) 185 

5.14 Internal Peak Temperatures During Curing 185 

5.15 Matrix for Multiple Linear Regression Analysis 205 

5.16 Equations of Fitted Curves from 1 st Regression Analysis 206 

5.17 Summary of 1 st Regression Analysis 206 

5.18 Equations of Fitted Curves from 2 n Regression Analysis 208 

5.19 Summary of Regression Analysis When 

Including the TB Interaction Term 208 

5.20 A Comparison of Equations of Fitted Curves From 2 n and 3rd Regression 
Analysis for Compressive Strength 212 

5.21 Comparing Significant Variables, R , and Standard Deviations for 
Compressive Strength 212 

5.22 A Comparison of Equations of Fitted Curves From 2 n and 3rd Regression 



xvi 



Analysis for Permeability 213 

5.23 Comparing Significant Variables, R , and Standard Deviations for 
Permeability 213 

5.24 A Comparison of Equations of Fitted Curves from 2 n and 3rd Regression 
Analysis for Length Change 214 

5.25 Comparing Significant Variables, R , and Standard Deviations for 
Permeability 214 

5.26 Summary of F-Statistic and P- Value from ANOVA 215 

6.1 Order of Performing Mixtures 227 

6.2 Base Mixture Design 227 

6.3 Phase I Testing Summary for Each Mixture 228 

6.4 Phase II Testing Summary for Each Mixture 228 



xvn 



LIST OF FIGURES 

Figure 

1.1 Concrete Consumption Forecast Compared Against Population Growth (Mehta and 
Monteiro, 2006) 2 

2.1 Location of Rajkot within the state of Gujarat, India (Google Maps) 19 

2.2 Paver Blocks (a) Removal from Molds (b) Design on Surface of Blocks 21 

2.3 Materials Stock Piled Directly on Construction Site 23 

2.4 Large Scale Used for Measuring Aggregate and Cement before Batching 24 

2.5 Materials Transferred from Scale into Portable Diesel Powered Mixer 24 

2.6 Laborers Placing Concrete 24 

2.7 Cement Being Emptied from the Bucket and Pulley Machinery 25 

2.8 Breakup of Landuse within City Limits of Rajkot (Rajkot Municipal Corporation, 
2006) 26 

2.9 Small Residential Buildings Near the Edge of City Limits 26 

2.10 Indoor Stadium 27 

2.11 Buildings Near the Center of the City 27 

2.12 Waste Water Treatment Plant 27 

2.13 Construction of Housing 28 

2.14 Construction of a Water Tower 28 

2.15 Tube Solar Water Heaters Mounted on the Roofs in Rajkot 30 

2.16 Rajkot Municipal Corporation Office with Passive Cooling Foyer Design 30 

2.17 Recharging Pit or Detention Pond Park Being Cleaned 31 

2.18 Park Filled with Stormwater After a Rain Event 32 

2.19 HVFA Concrete Road on Saurashtra University Campus (a) Two Wheelers and 
Tractor on the Road (b) Close up of the Surface of the Road 33 

xviii 



2.20 Raiya WWTP Site (a) Placing Concrete (b) Curing Concrete 33 

3.1 Life Cycle Phases and Material Flow for Concrete in Rajkot 44 

3.2 Trend in Cement Production for Four Leading Cement Producing Countries (USGS, 
2012; Parikh, Sharma, Kumasr, Vimal, IRADe, 2009) 47 

3.3 Potential Trend in Per Capita Cement Consumption for Four Leading Cement 
Producing Countries (USGS, 2012; Parikh, et al, 2009; United Nations 2010b) 47 

3.4 Steps in cement manufacturing process at Grasim Industries Limited Cement 
Company (Grasim Industries Limited, 2008) 53 

3.5 Phase Diagram for Ordinary Portland Cement (Gani, 1997) 55 

3.6 Cyclone Heat Exchangers and Precalciner (Gani, 1997) 60 

3.7 Indian Cement Emission Factors for 1991-2010 67 

3.8 Concrete Mixer with Mechanical Hopper 84 

4.1 Proposed Pervious Concrete System Site 93 

4.2 Pervious Parking Lot Pavement on Auraria Campus in Denver, Colorado 94 

4.3 Details of the Pervious Concrete System for the Parking Lot Installation (Hager, 
2009) 95 

4.4 Mixture Consistency (a) Too Dry, (b) Proper Amount of Water, (c) Too Wet (Tennis, 
Leming, & Akers, 2004) 99 

4.5 Compressive Strength Testing (a) Using Neoprene Pads for Cylinders and (b) Steel 
Plates for Cubes 101 

4.6 Hydraulic Testing Apparatus (a) Cylinder with Stopper and Putty (b) Hole Drilled in 
Cylinder for Draining Water from the Cylinder into the Pervious Concrete 104 

4.7 A Side by Side Comparison of the Pervious Concrete Samples 106 

4.8 Average Compressive Strengths for Mixture 1 and Mixture 2 110 

4.9 Fracture Paths for Cylinder Pervious Concrete Samples Ill 

4.10 Fracture Paths for Cube Pervious Concrete Samples Ill 

4.11 Fracture Occurring Through the Aggregate Ill 



xix 



4.12 Second Proposed Site for the Pervious Concrete System Placement 114 

4.13 (a) A Perforated Pipe Placed in Barrel (b) Image of Barrel 118 

4.14 Base and Sub-Base Layers a) Coarse Aggregate Layer b) Fine Aggregate Layer ..118 

4.15 Cloth Fiber used between Coarse and Fine Aggregate Layers 118 

4.16 Profile of the Pervious Concrete System Placed in the Barrel 119 

4.17 Evaluation of Pervious Concrete Consistency 121 

4.18 Rodding the Layers of Pervious Concrete in the Cube Mold 122 

4.19 Compacting the Pervious Concrete in the Cube Molds Using (a) Direction 1 and (b) 
Direction 2 122 

4.20 Covering the Pervious Concrete with a Wet Jute Bag 123 

4.21 Removal of Pervious Concrete from Cube Molds (a) Close-Up View (b) All Six 
Cubes 123 

4.22 Placing Pervious Concrete Cubes in a Water Bath 124 

4.23 Placement of the Pervious Concrete Samples in Water Filled Container to Determine 
Percentage Voids from Volume of Displaced Water 124 

4.24 Compressive Strength Test and Fracture Path 128 

4.25 Visual Observations (a) The Sample after Completion of Compressive Strength Test 
(b) Breaking the Sample Further by Hand 128 

4.26 Before and after Percolation (a) Bore Water Samples (b) Stream Water Samples .130 

4.27 Samples Collected for Pathogen and B.O.D. Tests (a) Bore Well Water Samples (b) 
Stream Water Samples 130 

4.28 Steel Roller for Compaction (a) Side View (b) Front View 136 

4.29 Sieve Analysis (a) Phase II Coarse Aggregate, (b) Rajkot Coarse Aggregate, (c) 
Phase II Fine Aggregate, (d) Rajkot Fine Aggregate 140 

4.30 Coarse Aggregate (a) Rajkot (b) Phase II 141 

4.31 Compressive Strength Fractures for M3 (a) Cubes and (b) Cylinders 144 



xx 



4.32 Compressive Strength Fractures for M4 (a) Cubes and (b) Cylinders 144 

4.33 Fracture Through Aggregate 145 

4.34 Average Compressive Strength of Cylinders and Cube Mixes for Pervious Concrete 
Designed for 2000 psi (13.8 MPa) Strength 146 

4.35 Relationship between Cylinder and Cube Average Compressive Strengths 147 

4.36 Average Compressive Strength with Standard Deviations for All Batches 148 

4.37 Average Compressive Strength with Standard Deviations between Cylinders and 
Cubes at 7-day and Final-Day Testing for all Batches 149 

4.38 Summary of Percent Porosity for All Batches 151 

4.39 Summary of Hydraulic Conductivity for all Batches 152 

4.40 Summary of Hydraulic Conductivity for all Batches Using Falling Head Criteria. 152 

5.1 (a) Vanakbori Fly Ash, (b) Gandhinagar Fly Ash 165 

5.2 Batches (a) Vanakbori and (b) Gandhinagar 167 

5.3 Cubes (a) Vanakbori and (b) Gandhinagar 167 

5.4 Average Compressive Strength Result for Rajkot HVFA Concrete Samples 168 

5.5 U.S. and India HVFA Concrete Average Compressive Strength Results 170 

5.6 Summary of Average Compressive Strength Results and Standard Deviations 
between the U.S. and Indian Sources of Fly Ash 171 

5.7 Aggregate (a) Storing and Cooling in a Shed and (b) Stockpiling 173 

5.8 Temperatures of Stock-Piled and Stored/Cooled Aggregate 175 

5.9 Campbell Scientific Datalogger (CR 10X) Used to Record Concrete 
Temperatures 177 

5.10 Installing the Thermocouple Into Concrete Sample 177 

5.1 1 Internal Curing Temperatures of Ambient Cured Fly Ash and OPC Samples 
During Trial 1 Testing 178 

5.12 Internal Curing Temperatures of Heat Cured HVFA and OPC Samples During 

xxi 



Trial 2 Testing 179 

5.13 Surface of Samples after Heat Curing (a) Fly Ash Mixture (b) OPC Mixture 179 

5.14 View of (a) Water Curing Tank (i.e. Ideally Cured) and (b) Hot Weather 

Curing Tank 182 

5.15 Hot Weather Simulation Tank (a) Boards to Keep Heat in (b) Close-Up of 
Aluminum Foil Bubble Insulation 183 

5.16 Schematic of Hot Weather Simulation Tanks 183 

5.17 Campbell Scientific (a) Datalogger (CR 5000) and (b) Setup for the Ideal and 

Hot Weather Simulation Tanks for Recording Concrete Temperatures 184 

5.18 Early Age Compressive Strength (a) No-Heated Aggregate (b) Heated 
Aggregate 188 

5.19 Later Age Compressive Strength (a) No-Heated Aggregate (b) Heated 
Aggregate 189 

5.20 Compressive Strength Results (a) No-Heated Aggregate, (b) Heated 

Aggregate 190 

5.21 Modulus of Elasticity (a) No-Heated Aggregate Concrete (b) Heated 

Aggregate Concrete 192 

5.22 Permeability Testing Setup 193 

5.23 Average Rapid Chloride Ion Permeability Test Results (a) No-Heated 
Aggregate, (b) Heated Aggregate 197 

5.24 Length Change Apparatus 198 

5.25 Length Change for No-Heated Aggregate Samples 200 

5.26 Length Change for Heated Aggregate Samples 201 

5.27 Effects of the Interaction of T and B on Compressive Strength 209 

5.28 Effects of the Interaction of T and B on Permeability 210 

5.29 Effects of the Interaction of T and B on Percent Length Change 211 

6.1 Sample Schedule for Competing Phase I-II Testing 228 



xxn 



6.2 Example of Finite Element Mesh and a Close-Up of a Single Element Based on 
Dimensions of the Length Change Beam Made in Lab 231 

6.3 Difference between Elastic Potential Energy of Water Cured and Heat Cured OPC 
Concrete Sample after 90 Days of Curing 232 

6.4 Schematic of Placement of the Thermocouple in Concrete Cylinder 234 



xxm 



1. Introduction 

1.1 Concrete and Urban Infrastructure 

With more than half of the world's population living in cities the demand of 
having effective and well functioning infrastructure for urban areas grows. Many 
governments identify the modernization of urban infrastructure as a crucial step for future 
economic growth and competitiveness. However, executing plans for infrastructure in 
any nation is a challenge because it usually involves long term strategies and allocating 
large amounts of funding even during times of fiscal strain. During the next forty years 
infrastructure is expected to cost approximately $70 trillion worldwide with most 
spending priorities occurring in the sectors of power/energy, residential, roads/bridges, 
rail, mining, healthcare, and water infrastructure (KPMG International, 2012; Seimens, 
GlobeScan, MRC McLean Hazel, 2007). 
1.1.1 Concrete Use 

Much of the urban built environment is constructed from the material known as 
concrete. Concrete is one of the most versatile construction materials next to steel. 
Concrete infrastructure has had a historic presence dating back to the rule of the Roman 
Empire and possibly originating 2000 years before the Romans during Egyptian times. 
Even today the basic ingredients within concrete are rock (coarse aggregate), sand (fine 
aggregate), water, and a cement powder (once the powder is mixed with water it acts as a 
binder for the rest of the ingredients). A typical concrete mix (portland cement concrete) 
usually consists of 15% cement by weight (FHWA, 2012). Cement production for the 
world between 2009 and 2010 was approximately 3310 million tonnes (USGS, 2012). 
Assuming the cement production is the potential consumption for the world and all 



cement is used for concrete products consisting 15% cement (by weight) then concrete 
consumption in 2009-2010 was about 22.1 billion tonnes. This estimate exceeds the 
forecasted concrete consumption that was presented by Mehta and Monteiro (2006) (See 
Figure 1.1). Mehta and Monteiro presented Figure 1.1 based on consumption rates 
leading up to the year 2002. Concrete consumption was estimated to peak at 16 billion 
tonnes (18 billion tons) or 2 tonnes/person when the population was about 10.4 billion 
people. With approximately 6.8 billion people between 2009 and 2010 per capita 
concrete consumption was about 3.3 tonnes of concrete/person. Note: At this time it 
seems as though no one entity keeps record of world consumption and production of 
concrete. Some countries or regions keep record of ready mix concrete use but in 
developing countries where concrete mixing occurs on-site this does not seem to be taken 
into account. 



O 

c 

o 

"^ 
S2 
=i 

CL 

o 



12 



11 



10 













Population 














/ s 






Concrete 


/ f 

// 
/t 






\ 

\ 


/f 






V 

\ 
\ 
\ 
\ 

V 

\ 



24 



2000 



20 



18 .9 



w 

o 



16 



Cl. 
t 
- 
X 

t= 
o 
u 

s 

14 I 

o 



12 



2025 



2050 
Year 



2075 



2100 



Figure 1.1 Concrete Consumption Forecast Compared Against Population Growth 
(Mehta and Monteiro, 2006) 



1.1.2 Concrete Infrastructure Is a Source of GHG Emissions 

Concrete infrastructure comes with an environmental price. Cement production, 
alone, can contribute between 5 to 7% of global green house gas emissions (Mehta and 
Monteiro, 2006; Crow, 2008). Concrete has a large carbon footprint not just due to 
cement, but because it is used in large amounts in urban areas as a construction material. 
As cities continue to grow, demand for new and maintained infrastructure intensifies, 
which leads to a continued release of greenhouse gas emissions. Greenhouse gas 
emissions have been tied to global warming or where the average weather is subject to 
warmer changes (one definition of climate change) through scientifically based 
assumptions. 
1.2 Climate Change in Urban Areas 

Scientific evidence points to urban areas as a major contributor to greenhouse gas 
(GHG) forced climate change. Anthropogenic (relating to influence of human activity) 
waste heat in the form of heating and cooling buildings, traffic, construction, and industry 
coincides with increasing urban heat islands and doubled carbon dioxide emissions. 
Climate model projections isolating the response of urban micro-climates to local 
(anthropogenic waste heat) and global effects, show that cities will experience an 
increase in maximum temperatures and frequency of hot nights. For cities such as Delhi 
or Los Angeles, 32 to 41 additional hot nights are a result of a low surface heat capacity, 
low soil moisture, high energy gains throughout the day, and rapid release of heat from 
the soil (McCarthy, Best, & Betts, 2010). 

However, a particular challenge in addressing climate change and (GHG) 
emissions is engaging cities to have a personal connection with climate change. Various 



cities across the world do not see climate change as being an issue, because there are 
uncertainties or skepticism regarding the cause and seriousness of climate change; 
climate change is believed to be too distant of a problem, or making changes are costly 
and undesirable for the current lifestyle, and climate change is not a priority compared to 
other issues requiring government action (Lorezoni, Cole, Whitmarsh, 2007). Globally, 
city governments are constantly balancing maintenance, design, and financial obligations 
to keep urban infrastructure reliable and safe. However, climate change can exacerbate 
these common infrastructure issues. In fact, growing research in sustainable 
infrastructure and climate action planning has provided evidence that failing to address 
the results of climate change and greenhouse gas emissions could lead to risks of 
increased water demand, heat island effect, declining air and water quality, new and old 
health risks emerging, and increased stresses and deterioration on the operation of urban 
and rural infrastructure (The World Bank, 2008; McCarthy, Best, & Betts, 2010, 
Mehrotra, Natenzon, Omojola, Folorusho, Gilbride, & Rosenzweig, 2009). Recently, 
urban areas are experiencing the effects of a changing climate. Some examples of 
present risks for infrastructure, in both developed and developing countries, based on 
current climatic conditions are described in the next few paragraphs. Note: The next few 
examples of climate change and current risks make reference to India and the United 
States because this dissertation has a focus on India. 
1.2.1 Flooding or Drought in Urban Areas 

Every year Indian cities experience flooding from seasonal monsoons that result 
in damage to urban infrastructure and increased health risks, however, urban 
infrastructure may have a contribution to flooding if inadequate stormwater control 



mechanisms are not developed or improperly developed. But for cities in arid to semi- 
arid regions any type of water is an important commodity to conserve. As will be 
explained in a later section this dissertation uses Rajkot, India as a case study for 
discussing urban concrete infrastructure opportunities to address carbon mitigation and 
climate adaptation. In a city like Rajkot, India the climate is hot and dry but flooding can 
occur with just rainfall intensities of 100 mm (4 in) due to nonexistent or few storm water 
management strategies and hard basaltic rock underlying the top soil of the terrain. But a 
concern other than flooding is the potential for climate change to produce more heat and 
lower rainfall. For the year 2012, in the state of Gujarat, India 14 districts and 152 
talukas (subdivision of a district) declared a state of drought due to receiving less than the 
average rainfall of 33 to 152 cms. Water sustains life in many of desert like regions but 
urban stormwater management solutions in countries such as India do not currently take 
into account the climate change risks for different regions. 
1.2.2 Extreme Temperatures in Urban Areas 

As mentioned previously extreme temperatures can become another concern for 
urban infrastructure. Road infrastructure is a priority for most countries. Currently more 
than 70% of paved roads in India are bitumen, with some major highways being concrete. 
However, there is increasing interest in investing in road projects using concrete. The 
Indian cement industry proclaims the benefits of concrete in road projects as "sound 
infrastructure," "long lasting," and can "save precious foreign exchange" spent on 
bitumen (CMA, 2010c). Bitumen for asphalt pavements has to be imported into India 
(CII, NRC, Ambuja Cement, 2004). The performance of concrete pavements has been 
successful in countries like the U.S. however in 201 1 the world experienced one of the 



warmest years on record leaving a reminder that climate change can affect the soundest 
of infrastructure. Sections of concrete pavement buckled due to trying to expand in 
weather that was consecutively above 32.2°C (90°F). The Minnesota Department of 
Transportation explained to Minnesota Public Radio News that concrete pavements had 
little room to expand near congested joints thus causing a lift to occur. In Oklahoma City 
a prolonged heat wave (+37.8°C [100°F]) caused concrete roads to buckle near joints 
similar to Minnesota, and the incidents were recorded across the entire state of 
Oklahoma. Research on climate change and concrete infrastructure has also shown that 
concrete infrastructure will face additional deterioration, carbonation, and chloride 
induced corrosion as a result of climate change events and increased greenhouse gas 
emissions (Wang, Nguyen, Stewart, Syme, & Leitch, 2010). 

Concrete has been identified as having a contribution to greenhouse gas emissions 
and also having susceptibility to damage and deterioration from the effects of changes in 
the climate. But the unique property of concrete, as stated previously, can be its 
versatility and ability to serve various purposes (such as climate adaptation) by adjusting 
the mixture design to include other materials aside from the four key ingredients. 
Additionally, if these materials can replace the use of cement then green house gas 
emissions can be reduced. 
1.3 Concrete Infrastructure for GHG Mitigation and Climate Adaptation 

If cities can identify an association with climate change then an important next 
step can be the assessment of effective and efficient adaptation or mitigation strategies 
and policies for urban areas and its infrastructure. Urban areas are complex systems and 
the vulnerability of each city depends on geographic, sectoral, and social attributes 



(Mehrotra et al., 2009). Many organizations such as the World Bank, Intergovernmental 
Panel on Climate Change (IPCC), Environmental Protection Agency (EPA), and 
International Council for Local Environmental Initiatives (ICLEI) are providing local 
governments with resources and ideas that can help urban areas prepare, prevent, or adapt 
to the possible effects of climate change. The study in this dissertation commenced with 
collaborative work between the University of Colorado Denver's Integrative Graduate 
Education and Research Traineeship program on Sustainable Urban Infrastructure and 
ICLEI-South Asia to develop sustainability assessments of infrastructure and develop 
decision support tools customized to Indian infrastructure (i.e. greenhouse gas (GHG) 
inventories that includes the building, transportation, construction material sectors) for 
cities in South Asia. 

Although cities may not know the exact vulnerabilities that urban areas and 
concrete infrastructure face under climate change and GHG emission increases it is 
expected that increasing GHG emissions leads to an increased risk of climate change 
occurring, and with climate change there is the likelihood that flooding, drought, and 
increasing temperatures (along with heat islands in urban areas) will have an influence on 
urban areas and infrastructure. This dissertation proposes that two concrete technologies 
exist to aid in climate adaptation and carbon mitigation for urban areas; pervious concrete 
and high volume fly ash concrete. 
1.3.1 Pervious Concrete Past and Contemporary Research 

Pervious concrete is known as a permeable, gap-graded, or porous concrete which 
allows water to percolate through intended voids in the concrete. A mixture design 
usually consists of higher proportions of coarse aggregate compared to conventional 



concrete, a thin layer of cementitious paste to bond and cover the aggregate, and little to 
no fine aggregate. In 1852 the United Kingdom began using the "no-fines" concrete (a 
form of the pervious concrete) as a construction material for buildings (Ghafoori & Dutta, 
1995). However, today pervious concrete is better known in the U.S. as a best 
management practice (BMP) technology because it can serve as a stormwater 
management tool that can recharge the groundwater, reduce stormwater runoff, reduce 
the level of contamination in run off, and help lower the heat island effect due to its open 
pore structure and its lighter color than asphalt pavements (Tennis et. al, 2004). Also, 
these same properties have led it to its description as a sustainable concrete. Research 
conducted at the University of Colorado Denver (UCD) revealed these various benefits in 
a pervious concrete pavement field installation (Hager, 2009). The successful installation 
involved the incorporation of 20% fly ash to offset the use of cement, 10% replacement 
of sand with crushed glass in the sub-base layer and the test section was monitored for 
deterioration, clogging, stormwater quality and reduction of the heat island effect. The 
results led to recommendations on design, placement and curing in order to produce 
durable pervious concrete pavements with sustainable aspects for urban areas in 
Colorado. Hager's research is one of many types of research exposing the benefits and 
promoting the use of pervious concrete. Between 2006 and 2009 research topics ranged 
from lab and field tests on pervious concrete to analyzing the capabilities of pervious 
concrete to filter compost effluent resulting from agriculture. The various types of 
research regarding pervious concrete can be found in appendix A Tables A. 1(a) through 
A. 1(e) which lists the research titles, authors, and objectives. Many of these studies have 



encouraged that cities use pervious concrete for other applications besides pavements and 
are listed below: 

• Alleys and driveways 

• Highway shoulders 

• Sidewalks 

• Low water crossings 

• Sub-base for conventional concrete pavements 

• Patios 

• Walls 

• Noise barriers 

1.3.2 High Volume Fly Ash Concrete Research with a Focus on Thermal Properties 
High volume fly ash (HVFA) concrete has been identified as incorporating more 
than 50% of fly ash by mass of total cementitious material into concrete (Malhotra & 
Mehta, 2008). In the 1980s Malhotra began testing HVFA concrete by using Class F and 
Class C fly ash. Using higher volumes of fly ash in concrete proved to give concrete 
improved mechanical properties and possess benefits such as those listed in Table 1.1 
(Giaccio & Malhotra, 1988; Malhotra & Mehta, 2008, American Coal Ash Association 
[ACAA], 2003; ACAA, 2002). The benefits of fly ash concrete have been taken beyond 
the physical, chemical and economic characteristics such that the use of fly ash is an 
indirect solution to green house gas (GHG) emissions and is a means for reducing energy 
use from cement manufacturing. In addition, the use of fly ash is associated with 
avoiding landfill, and reducing the overconsumption of virgin materials. 



Table 1.1 Summary of the Benefits of Fly Ash Concrete 



Benefits of fly ash concrete 



• High performance/high ultimate • Can compensate for fines not found in 
strengths some sands 

• Improved workability and • Lowers water demand 
flowability 

• Reduced bleeding and • Reduced concrete shrinkage 
segregation 

• Reduced heat of hydration • Reduces wear on delivery and plant 

equipment 

• Improved durability through • Increased resistance to sulfate attack, 
reduced permeability alkali-silica reactivity (ASR), and other 

forms of deterioration 

In one particular study performed at the University of Colorado Denver replacement of 
cement with 20% and 40% fly ash in concrete mixes reduced greenhouse gas emissions 
by 21% to 36%. The study was also unique in showing how per capita usage of cement, 
within the City and County of Denver boundaries, contributed to the city's total 
greenhouse gas footprint (Reiner, 2007). Reiner's work made it possible for cities like 
Denver to understand how the environmental impact of the conventional and fly ash 
concrete mixes could be quantified and compared with a combined life cycle assessment 
and material flow analysis. Also such information could be used as a tool for making 
decisions about the impacts we want future infrastructure to have. 

One particular characteristic noted from a literature review on HVFA concrete 
was the reason for incorporating it into concrete in the 1930s; fly ash was and has been 
used to reduce the heat of hydration in mass concrete (Malhotra and Mehta, 2008). In a 
study by Malhotra along with Rivest and Bisaillon (as cited by Malhotra & Mehta, 2008) 
several concrete monoliths (some made from HVFA and the others made from 100% 
cement) showed a difference in temperature of about 22°C (39.6°F) with the lowest 



10 



temperatures occurring in the HVFA concrete monoliths. Cements available today have 
such a high reactivity that a high heat of hydration is likely to occur even in structures 
with thicknesses less than 50 cm (-20 in). Although the properties of modern cement 
have improved, cement's characteristics can render a structure susceptible to thermal 
(excessive temperature differences between the concrete and the surrounding 
temperature) and drying shrinkage (contracting of hardened concrete due to loss of 
capillary water) cracking. These two types of cracking are especially a problem during 
hot weather concreting. Table A. 2, found in Appendix A summarizes just a handful of 
past research on fly ash concrete related to hot weather concreting applications or 
experimentations . 

Hot weather concreting means that precautions must be taken when concrete mixing and 
placing is occurring at temperatures above 32° C (90° F) or when concrete temperatures 
are somewhere between 25° C and 35°C (77° F and 95" F). Common solutions for hot 
weather concreting are the following (PC A, 2002). 

• Cool concrete materials before mixing 

• Schedule concrete placements to limit exposure, thus avoiding pouring during the 
hottest part of the day 

• Use chilled water or ice as part of the mixing water 

• Use of a Type II moderate heat cement 

• While curing use sunshades, misting, or fogging to limit moisture loss 

• Apply moisture -retaining films after screeding 

Studies on HVFA concrete have shown that thermal and drying shrinkage cracking are 
minimized in the concrete as a result of the properties of the fly ash (Malhotra & Mehta, 

11 



2008; Ravina, 1981; Mehta, 2002; and Senthil and Santhakumar, 2005). HVFA concrete 
lowers internal curing temperatures due to fly ash having a lower reaction compared to 
cement. Ravina studied lower percentages of fly ash in concrete, but both Ravina and 
Mehta express that hot weather concreting with fly ash decreases water demand during 
mixing. Also, high concrete temperatures have been shown to reduce strengths in 
concrete, however, both studies by Mehta (2002) and Ravina (1981) proved that fly ash 
concrete strengths were typically higher than a reference mixture made with ordinary 
Portland cement at later ages when both types of concretes were cured in hot 
temperatures. Other research has shown that the long-term performance of fly ash 
concrete have led to more durable structures that require less maintenance (ACAA, 
2002). 

Mehta (2002), Senthil and Santhakumar (2005) monitored the internal curing 
temperature of fly ash concrete and showed that fly ash can prevent thermal cracking. 
The study by Senthil and Santhakumar (2005) is one of the few studies where the mixture 
designs involved the use of blended cements from India. In India blended cements can 
consist of fly ash and cement or ground blast furnace slag and cement which are blended 
during the cement manufacturing process. The percentage of fly ash in the blended 
cement study by Senthil and Santhakumar was not specified, however the results revealed 
that the heat of hydration could be about 5°C (9°F) higher for the blended cements when 
compared to a general purpose cement and a high-strength cement. The surprisingly high 
heat of hydration may have been attributed to the fineness of the grinding, according to 
the authors; nevertheless the strengths were comparable to the high-strength cement 
mixture. Mehta's study emphasized that high volume fly ash concrete (with Class F fly 



12 



ash) would be most beneficial in keeping temperature increases under 30 C (54 F) and 
under such temperature maintenance thermal cracking was prevented for a foundation 
placed under warm and humid conditions. When 50% or more fly ash is utilized, the fly 
ash and the cement complement one another, such that some heat is generated from the 
presence of cement but part of the heat is concentrated on the acceleration of the 
pozzolanic reaction. 

Besides possessing beneficial properties for hot weather concreting HVFA 
concrete has other thermal properties that could be related to energy efficiency. The 
research by Bentz et al. (2010) was unique in the aspect of examining the thermal 
benefits of hardened fly ash concrete while the previous authors monitored temperatures 
of fresh concrete and then evaluated the mechanical properties after hardening. Although 
the mechanical properties were of importance to Bentz et al. the goal of the research was 
to evaluate the energy efficiency or insulative potential of high volume fly ash concrete 
for use in buildings (residential or commercial). Bentz, et al. did comment that the 
aggregates affected the thermal conductivity of the HVFA concrete; however, other 
research referenced in Table A. 2 did not make reference to aggregate effects. Thus, it 
may be beneficial to research the temperature of freshly mixed fly ash concrete as 
affected by temperature of aggregate. 
1.3.3 Main Goal and Knowledge Gaps 

The research regarding climate change and carbon dioxide should not be 
overlooked. The literature review and recent events have supported the idea that carbon 
is linked to climate change, and urban areas are facing a new challenge that could bring 
flooding, drought, and rising and prolonged temperatures. There is no doubt that the 



13 



climate change and carbon dioxide expose society to environmental and health risks. 
But, there is minimal research regarding the effects of carbon dioxide and climate change 
on the urban infrastructure that society depends on. 

Without proper infrastructure planning and designing, that takes into account 
climate change impacts, there is the possibility that new infrastructure could experience 
premature deterioration while the deterioration rate of older infrastructure could be 
exacerbated. Based on the literature review very few studies exist that explore how 
concrete infrastructure will be affected. However, the literature review did highlight the 
benefits that pervious concrete and high volume fly ash concrete could contribute towards 
climate adaptation. Despite the 80 plus years of research regarding both pervious 
concrete and high volume fly ash concrete many city governments are unaware of these 
benefits and therefore do not encourage the regular use of these two concrete 
technologies (Ghafoori and Dutta, 1995; Solis, Durham, Rens and Ramaswami, 2010). 

Studies by Hager (2009) and Reiner (2007) are great examples of how they used 
their research to demonstrate and improve on the advantages of pervious concrete and fly 
ash concrete. Recall, that the study by Reiner also indicated that fly ash use in concrete 
designs can reduce emissions resulting from cement and the manufacturing of concrete. 
In another study by Reiner along with Ramaswami, Hillman, Janson, and Thomas (2008), 
it was found that just by including the embodied energy of key urban materials such as 
concrete, quantification of per capita GHG emissions was improved for the city of 
Denver and became the benchmark from which the city could begin developing ways in 
reducing their emissions as whole or within certain sectors such as the design of 
construction materials. 



14 



Main Goal of Thesis: 

The main goal of the study is to evaluate pervious and HVFA concrete's contribution to 

carbon mitigation and climate adaptation in cities. 

Summary of Knowledge Gaps 

However, in order to support these recommendations the following knowledge gaps, 

which were identified from the literature review, are studied further and play a major role 

in this dissertation. 

• GHG emissions - Reiner's study was unique in quantifying the emissions from 
ready mix concrete operations in a city (2007). However, for cities that rely on 
on-stie mixing operations, such as in developing countries (i.e. India), are 
emissions comparable to those of where cities primarily use ready mix 
companies? 

• Pervious Concrete - There has been no research regarding the ability to transfer 
well-established and research supported pervious concrete designs to other 
regions having material differences. Research has indicated that size and shape 
of aggregate can change certain properties of the pervious concrete but it is 
unclear, if all materials differed (aggregate, water, cement), whether these 
changes drastically affect strength, porosity, filtration, and hydraulic 
conductivity all at once. 

• High Volume Fly Ash Concrete - It is already known that HVFA concrete is a 
well established solution to lowering the heat of hydration and preventer of 
thermal and drying shrinkage cracks during hot weather concreting. However, 
the literature on climate change has indicated that there is the likelihood of 



15 



extended periods where temperatures will rise above average temperatures. The 
question that research has not quite answered is how does do these prolonged 
high temperature events affect materials before mixing? How are fresh concrete 
properties affected after mixing when materials have been affected by the hot 
weather? Do current hot weather curing methods work for extended periods of 
high temperatures? Will hardened properties change dramatically when 
temperatures extend past 28 or more days of curing? The main question that has 
been left unanswered is whether fly ash has the capabilities of mitigating the 
effects of extended periods of heat even when required to cure for 56 to 90 days? 

1.4 Thesis Objectives 

The collaborative work with ICLEI South Asia and University of Colorado Denver 

presented the opportunity to study the knowledge gaps mentioned in the previous section. 

Thus the main objectives of this research were the following 

• This study applied the powerful tool of MFA-LCA to determine the 
environmental impacts of pervious and HVFA concrete compared to ordinary 
Portland cement (OPC) concrete in a developing country 

• In this study a comparative assessment of pervious concrete mixture designs for 
structural and environmental benefits across the U.S. and India was performed 

• In this study it was necessary to determine whether there are structural and 
durability benefits from HVFA in concrete mixtures when subjected to extreme 
hot weather conditions 



16 



1.5 Organization of Thesis 

1) In Chapter 2 the case study is introduced. The city of Rajkot, India is described 
through climate, cement consumption, stormwater management, and current construction 
with and without fly ash concrete. The city's current interest in climate adaptation and 
GHG mitigation is also discussed 

2) The role the cement industry plays in India's economy and energy consumption is 
discussed in Chapter 3. The manufacturing process as well as carbon mitigation 
strategies being implemented by the industry in India is emphasized. The chapter ends 
with a material flow and life cycle analysis of concrete for Rajkot, India but generally 
applicable to any city in the state of Gujarat. 

3) In Chapter 4 the methods and results of a small demonstration of a pervious concrete 
system that occurred in Rajkot, India is disclosed as Phase I of the pervious concrete 
project. This part of the study led to a concern over comparisons in strengths between 
cube and cylinder samples. As such Phase II is used to discuss the attempt at establishing 
a relationship between cubes and cylinder properties. 

4) In Chapter 5 Phase I of the HVFA fly ash study involves a comparison between typical 
fly ash properties in India and the U.S. and is used to discuss the importance of design 
and test of high volume fly ash concrete mixtures. Cubes and cylinders strength results 
are compared for the U.S. and India as part of Phase II. Phase III is used to identify the 
benefits of high volume fly ash concrete over ordinary Portland cement concrete when 
subjected to representative temperatures of hot days experienced throughout arid and 
semi-arid regions of India. The chapter describes a multiple linear regression analysis 
used to determine the effects of a variety of experimental conditions and compositions on 



17 



ordinary Portland cement concrete mixtures versus high volume fly ash concrete 

mixtures. 

6) The study ends with Chapter 6. A summary of the major findings are discussed as 

well as recommendations on how to improve on the study. 



18 



2. Case Study Location: The City of Rajkot India 
2.1 Demographics, Population, and Climate 

Rajkot is located in the state of Gujarat in Western India (Refer to Figure 2.1). 
The climate of Rajkot is hot and dry throughout much of the year thus representing a 
semi-arid region. Mild temperatures can be about 20°C (68°F) but during the summer, 
during the months of March through June, temperatures range between 24°C to 42°C 
(75.2°F to 107. 6°F). Rajkot can experience acute droughts at times but, during the 
monsoon period (June to September) the city can receive an average of 500 mm (19.7 in.) 
of rain. 




Figure 2.1 Location of Rajkot within the state of Gujarat, India (Google Maps) 



19 



The population of Rajkot according to a 2001 census was approximately 
1,002,000 and has increased to about 1.4 million. A growth rate of 79.12% was 
established in the 2001 census for the years (1991 to 2001) but was partly attributed to 
extending the city limits to include other villages. Rajkot city is connected to other parts 
of the country by air, two railway stations, and major roads that link Rajkot to several 
cities within the state including to the state capital Ghandinagar. Rajkot is considered an 
industrial town and the economy is based on over 400 foundries, engine oil 
manufacturing, machine tools, engineering and auto works, castor oil processing, jewelry, 
handicrafts, clothing, medicines, and agriculture (Rajkot Municipal Corporation, 2006). 

A combination of Rajkot's average growth rate of 3% and Rajkot's identity as an 
economic, industrial, and educational center has led to continued urbanization and the 
need for a comprehensive development plan. Rajkot Municipal Corporation city 
development plan for the years 2005-2012 was developed under the Jawahar Nehru 
National Urban Renewal Mission (JnNURM) such that the goal of the city was identified 
as being responsive, economical, efficient, productive, and equitable. While under the 
mission of the JnNURM, Rajkot has also committed to incorporating clean development 
strategies so that infrastructure investments would lead to an improved urban 
environment and sustainable city. Recognizing JnNURM's mission ICLEFs (Local 
Governments for Sustainability), South Asia Urban Climate Project has been working 
with Rajkot to begin implementing sustainable infrastructure interventions that address 
the infrastructure problems identified in the city development plan. 



20 



2.2 Rajkot Construction and Concrete Infrastructure 

At least 5.5% of Rajkot's population is employed through the construction 
industry. The highest employment (28% of total population) occurs within the sector of 
manufacturing. Rajkot does not have a cement manufacturing plant within city limits. 
The closest plant is located about 116 km (72 mi) outside of the city in the area known as 
Sikka. According to the Cement Manufacturers' Association of India 7 other large 
cement manufacturing plants are located within the state of Gujarat and the farthest plant 
about s 295 km (183 mi) from Rajkot. Cement in Rajkot is used for various construction 
materials such as reinforced cement concrete, prestressed concrete, paver blocks, cement 
blocks, and asbestos piping. Figure 2.2a and 2.2b shows an example of paver blocks 
made in Rajkot. 




(a) (b) 

Figure 2.2 Paver Blocks (a) Removal from Molds (b) Design on Surface of Blocks 

Major cement companies that deliver cement throughout the state of Gujarat are Hathi 

Cement (part of SaurashtraCement Limited), Gujarat Sidhee Cement Limited, UltraTech 

Cement Limited (part of the Aditya Birla Group), Ambuja Cements Ltd., Shree Digvijya 

Cement Co. Ltd., HMP Cements Ltd, Sanghi Industry Ltd., JK Lakshmi Cement Ltd., 

and Jaypee Cement (CM A, 2010c). There is one ready mix concrete plant within the city 

limits which is owned and operated by the cement manufacturing company known as 



21 



Lafarge. Ready mix concrete, in Rajkot, is only used in large construction projects. 
Large construction projects are associated with bridges, fly overs, railway, skyscrapers, 
and bus rapid transit system (BRTS) roads (M. Joshi and Mr. Girish [contractor] personal 
communication, March 8, 201 1). The majority of the construction seen in Rajkot used 
the method of on-site mixing. 
2.2.1 Personal Account of Construction 

Collaborative work with Rajkot Municipal Corporation allowed for personal 
observations and communications to be made with a city assistant engineer and city civil 
engineer as well as a structural engineer/owner of Lakhlani Associates. Additionally the 
collaborative work allowed for the majority of the field research, presented in this 
dissertation, to be performed on-site where a water/tower (designed by Lakhlani 
Associates) was being constructed. The construction process of the water tower revealed 
the following about most of the city concrete construction projects: 

• Ready mix is expensive compared to on-site mixed concrete and is not considered 
necessary for all city projects 

• Cement bags and aggregate are delivered in bulk to the site (Refer to Figure 2.3 
for example of stock piled materials) 

• Most common cement used on-site was Hathi, Ambuja, UltraTech, and Sidhee 
cements 

• At this particular site water used for concrete mixture design was taken from a 
bore well drilled on site 



22 



• Each batch required that aggregate be weighed using a large scale located on site 
that was calibrated daily. (Refer to Figure 2.4) 

• Concrete was mixed with portable diesel powered commercial concrete mixers 
(See Figure 2.5) 

• Mixed concrete was transported by wheelbarrows or up several heights by a 
bucket and pulley (See Figure 2.6 and Figure 2.7). 

• Bamboo was used for scaffolding and concrete forms 

• Both men and women worked and lived on-site 

• Most of the laborers were from villages nearby 

• Not all laborers had safety equipment to wear. 

• The laborers who worked with the placing of steel reinforcement are considered 
skilled workers and get paid more than those working with just concrete 




Figure 2.3 Materials Stock Piled Directly on Construction Site 



23 








>U 



mmS 




i 




,^*k 



Figure 2.4 Large Scale Used for Measuring Aggregate and Cement before Batching 




Figure 2.5 Materials Transferred from Scale into Portable Diesel Powered Mixer 




Figure 2.6 Laborers Placing Concrete 



24 




Figure 2.7 Cement Being Emptied from the Bucket and Pulley Machinery 
2.2.2 Rajkot Concrete Infrastructure 

In 2001 the city of Rajkot occupied an estimated 10,485 hectares (25906 acres) of land. 
Figure 2.8 displays the breakup of land use in Rajkot. About 74% of the city limits were 
developed, with residential areas occupying a little more than half of the developed (i.e. 
residential, commercial, industrial, transportation, public, recreational, and other) area. 
Commercial use is mostly reserved for retail marketing, industrial use includes 369 units 
of various industries within the city limits and public use include hospitals, schools, and 
government office buildings. 

Investment on infrastructure projects in Rajkot occurs in the sectors of traffic and 
transport, water supply, drainage, stormwater drainage, housing and the urban poor, 
public works, and solid waste management. The majority of built infrastructure is 
constructed of concrete. The typical concrete infrastructure seen in Rajkot can be 
described as follows and are depicted in Figures 2.9 through 2.14: 

• Recreational/Office/Home/ Apartment Buildings 

• Roads 

• Sidewalks 



25 



• Wastewater Treatment Plants 

• Check dams 

• Water piping systems 

• Bridges 

• Water Towers 

Total Land = 10485 Hectare 



Water Bodies 

2% 



RecreationaL 
Space 

1% 



Public and Semi_ 
Public 

1% 




Traffic and 

Transportation 

13% 




Industrial 



Commercial 

2% 



1 Hectare = 2.471 Acres 

Figure 2.8 Breakup of Landuse within City Limits of Rajkot 

(Rajkot Municipal Corporation, 2006) 




Figure 2.9 Small Residential Buildings Near the Edge of City Limits 



26 




Figure 2.10 Indoor Stadium 




Figure 2.11 Buildings Near the Center of the City 




Figure 2.12 Waste Water Treatment Plant 



27 




Figure 2.13 Construction of Housing 




Figure 2.14 Construction of a Water Tower 

Concrete plays a major role for the infrastructure found in Rajkot. Despite on-site mixing 
seemingly lagging in terms of modern construction, Indian structural engineers have been 
successful in demonstrating the advantages of concrete design for structures. The indoor 
stadium was designed by Lakhlani Associates and in 2005 Mr. Lakhlani received the "R 
H Mahimtura Award For Excellence In Strucutral Engineering" due to the innovative 
structural system designed for the stadium. One unique aspect that made the design 
innovative was a reinforced concrete tripod system which has the purpose of transmitting 
roof forces to the ground. The stadium is just one example of how Rajkot has a 
distinctive type of city management that is eager to collaborate with private and 
government entities to try new ideas that allow the city to advance in terms of 



28 



technology, business, industry, and infrastructure. In fact the city development plan for 
2005-2012 expresses Rajkot's goals in developing infrastructure that will reduce GHG 
emissions and energy consumption and achieve a productive, efficient, equitable, and 
responsive city which is part of the Government of India's Jawaharlal Nehru National 
Urban Renewal Mission (JnNURM). 
2.3 Future GHG Mitigation and Climate Adaptation Goals 

Rajkot has already demonstrated how the goals of the city development plan are 
being achieved. On the roofs of many buildings in Rajkot, evacuated tube solar water 
heaters have been installed (See Figure 2.15). This type of device has been used for some 
years and is currently cheaper than gas water heaters; consequently it avoids CO2 that 
could result from gas powered water heaters. At the Rajkot Municipal Corporation 
western zone office a solar photovoltaic system was installed to partially power the 
office. Also, one of the Rajkot Municipal Corporation offices was designed with an open 
foyer so that the various floors were cooled through passive cooling. Figure 2.16 shows 
the open foyer which included a nice landscaping of plants. Other interventions that were 
being implemented during 201 1 was the construction of a Bus Rapid Transit System, 
solar powered lights for parks, investing in energy saving technologies for schools, and 
installing more city trash bins in communities throughout the city for waste collection. 
Some of the interventions were a result of the collaborative work with ICLEI South Asia. 
The collaboration was meant to showcase clean and efficient technologies for 
infrastructure and to find what could be successful for the city as a long term method of 
use or design. Rajkot has been willing to implement new ideas into their infrastructure 
design even before collaborating with ICLEI South Asia. As stated in the city 



29 



development plan some of the priorities of the city have been improvement of roads and 
stormwater management. The city has experimented with stormwater management 
methods and fly ash concrete roads. 




Figure 2.15 Tube Solar Water Heaters Mounted on the Roofs in Rajkot 




Figure 2.16 Rajkot Municipal Corporation Office with Passive Cooling Foyer 
Design 



30 



2.3.1 Stormwater/Rainwater Harvesting 

Through a personal communication with Y. K. Goswami (March 9, 2011), 
Assistant Engineer at Rajkot Municipal Corporation, Rajkot has been involved in urban 
rainwater harvesting or stormwater management trials. For example within certain 
residential areas, parks or gardens have been built such the park acts like a recharging pit 
or detention pond when it rains. In Figure 2. 17 an example of one of these parks can be 
seen before it is filled by stormwater. The purpose of these parks is to direct the 
stormwater into these pits so that the water either seeps into the ground or in some cases 
drains into a storage tank below constructed below the park. The depth excavated for 
these parks will vary based on the rain events expected for an area. In Figure 2. 17 it 
appears as though the depth is at least 1.2 m (4 ft). Figure 2.18 shows the same park after 
storm water had drained into the park. The rain event filled the total depth of the park. 




Figure 2.17 Recharging Pit or Detention Pond Park Being Cleaned 



31 




Figure 2.18 Park Filled with Stormwater After a Rain Event 
2.3.2 HVFA Concrete Road Project 

In 2005 Rajkot finished construction of the city's first fly ash concrete road. The 
project was completed in partnership with Ambuja Cement, Natural Resources of 
Canada, and the Confederation of Indian Industry. The project used a high volume fly 
ash concrete mixture design. The mixture design demonstrated that initial costs for 
concrete roads could be reduced through the use of local materials and waste products 
such as fly ash. The project presented an alternative to bituminous roads. The road 
extends 2.3 km (1.4 mi) through the campus of Saurashtra University in Rajkot. The 
material design included grade 53 ordinary portland cement (OPC) from Ambuja 
Cements Ltd. and fly ash from Sikka thermal power plant located in Sikka, Gujarat, 
which is about 115 km (71 mi) from the city of Rajkot. Two concrete layers made up the 
design of the road, such that the 150 mm (6 in) thick bottom layer was from a 50% high 
volume fly ash concrete mixture and the 50 mm (~ 2in) thick top layers was made from a 
30% fly ash concrete mixture. The top and bottom layer reached a compressive strength 
of about 41.2 MPa (5976 psi) and 40.1 MPa (5816 psi) respectively. Design compressive 
strengths for concrete roads in the U.S. generally are at least 28 MPa (4000 psi) and 

Rajkot' s road project was at least 12 MPa (-1800 psi) greater than the design 

32 



compressive strength. Figures 2.19a and 2.19b show the surface of the concrete 
pavement 6 years after it had been constructed. Although Figure 2.19a shows a two- 
wheeler and tractor using the road, heavier traffic such as commercial vehicles can be 
expected on the road as well. Figure 2. 19b shows the different wearing down of the 
surface of the concrete. The project brought about other concrete pavement construction 
and in 201 1 a high volume fly ash concrete road was being constructed around the Raiya 
waste water treatment plant in Rajkot (See Figures 2.20a and 2.20b). 












-' ''7-V-- ■ '-'. 
















***7'£ -*■'"» 






. *.■&! 








'£% 


' j *- ""*, " ' " 


'* * v ■ V 








*t, 




■Cr'M • 











(a) (b) 

Figure 2.19 HVFA Concrete Road on Saurashtra University Campus (a) Two 
Wheelers and Tractor on the Road (b) Close up of the Surface of the Road 





(a) (b) 

Figure 2.20 Raiya WWTP Site (a) Placing Concrete (b) Curing Concrete 



33 



2.3.3 Collaboration between UC Denver, ICLEI South Asia and Rajkot Municipal 
Corporation 

As part of Rajkot' s collaboration with UC Denver and ICLEI South Asia it was 
decided that there was further interest in trying new stormwater management systems or 
best practices and the need to further study the compressive strength with high volume fly 
ash concrete. Originally all parties preferred an actual field installation of a pervious 
concrete system and anticipated the results for compressive strength, water percolation, 
and water quality improvements. The parties agreed that the pervious concrete field 
demonstration would be constructed at the Raiya WWT site where the fly ash concrete 
road was being placed. As part of the fly ash concrete project there was significance 
placed on determining whether other local fly ash sources (besides the Sikka power plant 
fly ash used in Saurashtra University road) would produce similar compressive strengths. 
There was an overall interest in promoting the use of these concrete technologies to 
facilitate reforms and improvement of urban infrastructure for cities interested in climate 
adaptation and carbon mitigation (carbon mitigation through quantification of reduced 
GHG emissions from use of the pervious concrete and fly ash concrete). The remainder 
of this dissertation will discuss the collaboration between parties in detail. A material 
flow and life cycle analysis (MFA-LCA) of cement and concrete will be discussed in 
Chapter 3. The MFA-LCA was modeled after the study conducted by Reiner (2007) with 
the goal of determining the contribution that cement use had in cities such as Rajkot. 
Chapter 4 provides the discussion of the potential applications of pervious concrete in 
Rajkot for stormwater management while Chapter 5 discusses the potential for HVFA 



34 



concrete to be used as a climate adaptation strategy in extreme hot weather conditions 
that could occur in a semi-arid region like Rajkot. 



35 



3. Carbon Mitigation Through Concrete: An MFA-LCA Approach 

3.1 Bottom-line: Cement and Concrete Manufacturing in India and the US 

The objective of this study was to quantify CO2 emissions resulting from Indian 
cement manufacturing and concrete production and compare results to the U.S. 
Literature reported different cement emission factors ranging from 0.6 to 1.0 tonne of 
C0 2 /tonne of cement (0.6 to 1.0 lb C0 2 /lb cement) (e.g. WBCSD, 2010; Parikh, Sharma, 
Kumar, Vimal, IRADe, 2009). It was unclear which would be the most appropriate 
emissions factor. Thus initial findings resulted in the review of Grasim Industries 
sustainability report published for the year 2007-2008. Grasim, ACC Ltd. and Ambuja 
Cements Ltd. are major competitors in Indian cement manufacturing. Both Grasim and 
Ambuja are providers of cement products to the state of Gujarat. Grasim's report was 
also one of the only available reports that had created a CO2 emissions and energy 
inventory that could be compared to the thorough inventory published for the U.S. 
cement industry by the Portland Cement Association (Marceau, Nisbet, VanGeem, 2010). 

Grasim's report presented a consolidated (including subsidiary companies) 
account of total materials, energy, and electricity used in the year. In addition, CO2 
emissions for direct energy (thermal energy) and indirect energy (purchased electricity) 
were calculated. Grasim's report indicated that all cement manufacturing plants had been 
converted into the dry precalcination process. There are three main processes for 
manufacturing cement and each are discussed later in this chapter. However, the dry 
precalcination process is currently the most energy efficient process available for cement 
manufacturing (about 1.3 GJ/tonne of clinker [559 Btu/lb clinker] more efficient than the 



36 



wet process). The precalcination process makes use of the waste heat from the kiln and 

clinker cooler to preheat the kiln material by use of cyclone preheaters installed before 

the kiln (up to 6 cyclones can be installed). Both in the U.S. and India the dry process is 

used to produce more than half the total cement produced, however the dry process in the 

U.S. only accounts for 53% of total production, but in India it is 98% (Maceau et al., 

2010 and CMA, 2010c). Table 3.1 compares the direct and indirect energy consumption 

for India (represented by Grasim) and the U.S. through the dry precalcination process. 

Table 3.1 Comparison of Energy Use per Tonne of Cement Between the U.S. 
Cement Industry and India's Grasim Industries. 



Energy Source 



Coal 
Gasoline 



Dry Precalcination Process 
GJ/Tonne of Cement 



U. S. India (Grasim) *India (Grasim) 



S? 

a 

Q 



PJ 
I 

a 




Liquefied Petroleum Gas 

Middle distillates 

Natural gas 

Petroleum coke 

Residual oil 

Wastes 

Furnace Oil 

Diesel 

Lignite 



Purchased Electricity 






Total 4.2 

* Values are in GJ/Tonne of cementitious material 
1 GJ/Tonne of cement = 429.92 Btu/lb of cement 
Source: Marceau, Nisbet, VanGeem, 2010; Grasim Industries Ltd, 2008 

An overview of an LCA is given in the next section, but a key step in an LCA is 

choosing a functional unit. In order to relate inputs and outputs of the cement 



37 



manufacturing process from two different countries the functional unit has to be the 
same. The functional unit for the LCA (cradle-to-gate) given in Table 3.2 was a unit 
mass (i.e. tonne) of cement. It is important to note that Grasim's report actually reported 
final emissions in terms of tonnes of CCVtonne of cementitious material. "Cementitious" 
is used to represent the use of alternative materials that are used in replacement of a 
percentage of cement. These materials can be fly ash, silica fume, or slag. In the U.S., 
the use of these materials is usually called blended cements (Type IP, Type IS, Type 
I(PM), and Type I(SM) where P = pozzolana, S = slag, M = modified) and in India these 
cements are called portland pozzolana cement (PPC) (when fly ash is used) and portland 
blast furnace slag cement (PBFS). Blended cement production in the U.S. is about 2 to 
3% of total production while in India it is about 60 to 70% (USGS, 2010; CMA 2010c). 
Use of these cementitious materials ideally reduces cement clinker demand for a unit 
mass of cement product as a result of less kiln fuel being burnt. Additionally, use of 
cementitious materials avoids disposal or stock piling of fly ash and slag. However, 
emissions do arise from transportation of the fly ash and slag to the cement 
manufacturing site and additional emissions may occur from any grinding that is 
necessary for slag. Since India produces large amounts of cementitious materials 
including it as the functional unit in a life cycle inventory as Grasim did is a benefit. 
However, it is not necessary because if there has already been a reduction in thermal and 
electrical energy due to less clinker is being processed this would be reflected in the 
inventory without using cementitious as the functional unit. 

In Table 3.1 direct energy for the U. S. and Grasim does not always come from 
the same fuels. According to Grasim's sustainability report fuels such as gasoline and 



38 



natural gas are not used as they are in the U.S. However, the coal consumption appears 

to be similar and there is about a 20% difference in the use of petroleum coke. Often use 

of alternative waste fuel materials (which are discussed in detail later in this chapter) in 

the kiln reduces carbon dioxide emissions. In this case, the U.S. cement industries on 

average use more waste materials as kiln fuels compared to Grasim. For indirect energy, 

there is a large difference in electricity purchased between Grasim and the U.S. Grasim 

uses about 65% less purchased electricity compared to the United States. In India captive 

power plants generate electricity on-site and reduce the need to purchase electricity from 

state grids. Overall Grasim uses approximately 10% less energy in the manufacturing 

process compared to the average cement industries in the U.S. that use the precalcination 

process. 

Table 3.2 Summary of Energy Use and Emission Factors from Direct and Indirect 
CO2 Emissions between India and the U.S. 



Attribute 


U.S. 


India (Grasim)* 


India (Grasim)* 


Functional Unit 


Cement 


Cement 


Cementilious 


Thermal Energy Use 
(GJ/tonne of cement) 




3.6 


3.0 


Purchased Electricity Use 
(kWh/tonne of cement) 


144 


51 


42 


^Electricity EF 
(kg C02/tonne of cement) 






36 


**Cement EF 
(kg C02/tonne cement) 


867 


855 


708 



* India has smaller emissions from electricity due to use of captive power on-site 

**This number is net electricity purchased, however, with the inclusion of indirect 

emissions this leads to about a 7% increase in emissions for the U.S. cement 

manufacturing. 

1 GJ/Tonne of cement = 429.92 Btu/lb of cement, 1 kWh/Tonne of cement = 1.54 Btu/lb, 

1 kg/tonne = 2 lb/short ton 

Table 3.2 uses the data from Table 3.1 to calculate electricity and cement (net 

purchased electricity) emission factors. In addition, the emission factors reflect how 



39 



using different fuels and different methods of attaining electricity can change the result of 
the emissions. Grasim's captive power plants, fuels, and use of cementitious materials 
help the industry reduce cement emissions by about 12 kgCCVtonne of cement (24 
lbCCVshort ton of cement) or about a 2% reduction. If purchased electricity was 
included in the cement emission factor the reduction is greater for Grasim, about a 7% 
percent difference. Also it might be important to note that if the functional unit was 
cementitious materials than Grasim shows a larger reduction in emissions. 

As stated previously the concrete emission factor for India was also important. 
Currently no published research could be found regarding an emission factor for concrete 
in India. In the U.S. two studies have reported a thorough inventory for concrete 
production. The study by Reiner (2007) discusses two different methods used to 
calculate a concrete emission factor. Using the software program Building for 
Environmental and Economic Sustainability (BEES) Version 3.0 Reiner estimated a 
concrete emission factor to be 0.17 tonnes CCVtonne concrete. Reiner's study improved 
on the BEES estimated concrete emission factor through the development of a life cycle 
analysis for concrete used in the city of Denver, Colorado. Through his LCA the 
concrete emission factor for a common type of concrete used in Denver (Class B) was 
estimated to be 0.22 tonne CCVtonne of concrete. Reiner demonstrated that the concrete 
emission factor will vary due to the reality of different concrete mixture designs. The 
second study, contracted out by the U.S. Department of Energy (2003), calculated a 
concrete emission factor equal to about 0.15 tonne COi/tonne concrete. 

Both studies by Reiner and the U.S. Department of Energy quantified concrete 
emission factors by gathering data from cement, aggregate, transportation, and ready 



40 



mixed operations. In India ready mixed concrete operations are not the commonly used 
method to produce concrete. As will be discussed in this chapter site-mixed concrete is 
the main method of producing concrete in India. Throughout this chapter, the 
development of a LCA (cradle to gate) for concrete will be discussed. Although 
Grasim's sustainability report presented a reasonable accounting of CO2 emissions it was 
decided by the author that a cement emission factor should be calculated to represent the 
majority of the companies that provide cement to concrete construction projects in 
Rajkot. The remaining sections in this chapter will discuss the energy consumption and 
efficiency methods being used by Indian cement manufacturing, the energy and CO2 
emissions for aggregate processing, transportation of materials, and on-site mixing of 
concrete. Finally a concrete emission factor will be calculated for a conventional 
concrete mixture used in Rajkot, and for pervious and high volume fly ash concrete 
mixture designs in order to show the environmental advantages of using pervious 
concrete and high volume fly ash concrete. 
3.2 Life Cycle Assessment of Cement and Concrete in India 

Tools such as an environmental life cycle assessment (LCA) can be used to assess 
certain environmental impacts (i.e. GHG emissions) that are associated with the different 
phases of a material or product. An LCA tool can also be applied as a strategy for 
determining how GHG emissions can be reduced to moderate the impacts of climate 
change. GHG are gases that trap heat in the atmosphere and are represented by global 
warming potentials (GWP) in CO2 equivalents (CC^eq). Carbon dioxide (CO2) is the 
baseline and has a global warming potential of 1; methane (CH4) has a GWP = 21; and 
nitrous oxide (N 2 0) has a GWP = 310 (fPCC, 2007a). Chlorofluorocarbons (CFCs), 



41 



hydrofluorocarbons (HFCs), hydrochlorofluorocarbons (HCFCs), perfluorocarbons 
(PFCs), and sulfur hexafluoride (SF6) have high GWPs and range from 90 to 23,900. 

The use of concrete in urban areas of India (i.e. Rajkot) is the focus for this 
dissertation. In this chapter CO2 impacts are quantified for concrete such that the 
boundary of the LCA begins with the manufacture of cement up to the concrete 
production method used most commonly in cities in India (i.e. site-mixed concrete). 
Insufficient literature exists on site-mixed concrete, but this study will be one of the first 
to apply the method of LCA to site-mixed concrete. In addition to the development of an 
appropriate LCA model this chapter will show the CO2 impacts of urban structural 
concrete mixture designs. These impacts will be compared to those calculated for 
previous concrete and high volume fly ash concrete to demonstrate the reduction in CO2 
that can be expected with the two sustainable concretes. 

An LCA takes into account energy, material inputs, and environmental releases 
from material acquisition, product manufacturing, transportation, use, maintenance, and 
disposal and/or recycling. In this dissertation the impacts from a cradle-to-gate (resource 
extraction to the product leaving the manufacturing process) and the product use phase 
are quantified. The end of life of the product is not considered in this study. 

In a LCA study on high performance concrete Reiner (2007) describes three 
different models (process based, economy input/output [eio], and hybrid [combined 
process and eio]) used to complete an LCA. This dissertation uses the process based 
LCA model such that the inputs (materials and CO2 energy resources) and the outputs 
(CO2 emissions) are itemized for producing concrete in India. The following 
methodology was used to complete the LCA model: 



42 



• Goal and scope defined -India is one of the largest producers of cement, 
however, not all the cement companies follow the same protocol for 
determining CO2 emissions. Therefore, within this study the process-based 
LCA model will be used to quantify CO2 emissions from fuel consumed in the 
cement manufacturing process. Calcining emissions that occur at the kiln will 
also be taken into account. The emissions from electricity are included from 
cement manufacturing as well. For the first the time, an emission factor will be 
calculated for the production of site-mixed concrete. This will also involve the 
emissions from fuel consumption during crushing of aggregate, transportation 
of aggregate, transportation of cement, transportation of fly ash, and operations 
of the portable cement mixer. The functional unit is one tonne of concrete 

• Inventory analysis - A description of the materials and processes used to make 
concrete is described throughout this chapter and the system boundary is shown 
in Figure 3.1 

• Impact Assessment - The only greenhouse gas taken into account for the 
quantification of emissions is CO2. The other five gases that can contribute to 
GHGs (methane, nitrous oxide, hydrofluorocarbons, sulfur hexafluoride, and 
perfluorocarbon) will not be included in this initial emissions study for Indian 
concrete. Normally these gases would be taken into account but there is limited 
information on these gases for cement manufacturing in India and carbon 
dioxide emissions are usually more significant than the emissions from methane 
and nitrous oxide. 



43 



• Interpretation step - The development of a model that can be used to determine 
the environmental impact of using a certain concrete mixture design will be 
useful in understanding which alternative mixture designs can provide the same 
serviceability and durability as well as reduce the environmental impacts. 



Material Extraction 
and Processing 



C one ret e Pro c e ssing 



Cradle-to-Gate 





/ 
1 




Fine and Coarse 
Aggregate 




Qn-site Concrete 
Mixing 





Cement 



Fly Ash 



id 






Rajkot City 



*(Figure design adapted from Reiner, 2007) Dotted line represents the city's 
environment, while the arrows represent the transportation used throughout the entire 
flow of materials. 
Figure 3.1 Life Cycle Phases and Material Flow for Concrete in Rajkot 

3.3 Understanding the Cement Production and Concrete Industry in India 

The production of cement is about a century old in India. The first cement 
industry was established in Porbundar, Gujarat in 1914 (DRPSCC, 2011). Table 3.3 
shows how India compares within the top 19 cement producing countries/regions in the 
world for the year 2010. In 2010 India was the second largest producer. In Figures 3.2 
and 3.3 the trend in cement production and potential cement consumption for four major 
cement producing countries are shown. Japan and India are unique in Figures 3.2 and 3.3 
because Japan currently has the highest kiln capacity (3370 tonnes per day) and is first in 
energy efficiency while India is second in both categories. Despite, India being a major 



44 



producer of cement the national per capita consumption is 0. 136 which is lower than the 
world average which is about 0.48 and lower than that seen in China, U.S., and Japan 
(Refer to Figure 3.2 and Table 3.3). However, India being one of the most populous 
countries and having an increasing economy have led the cement industry analysts to 
state that the slow increase in per capita cement consumption is just an indicator of the 
industry's growth potential (Ernst & Young, 201 1). According to the Cement 
Manufacturers' Association (CMA, 2010c) India has approximately 142 large cement 
plants together producing at least 161 million tonnes of cement a year. Two major 
companies ACC Ltd. And Ambuja Cements Ltd. withdrew membership from CMA thus 
there production is not included in CMA's statistics presented in Table 3.4. In fact 
during the 2009-2010 year Ambuja produced 20.1 million tonnes of cement while ACC 
Ltd. produced 21.4 million tonnes. The Indian cement industry has three types of cement 
units which are large, white, and mini cement plants. The mini plants use vertical shaft 
kilns with cement production not exceeding 109,500 tonnes/year (120701 ton/year [ton = 
U.S. short ton]) and are plants that use the rotary kiln such that cement production does 
not exceed 300,000 tonnes/year (330690 ton/yr). Within this research the focus is 
pertaining to large cement plants. 

According to an article in the Indian Concrete Journal, concrete was identified as 
the preferred construction material in India (Kumar & Kaushik, 2003). Between the 
years 1998-2003 major construction projects that utilized concrete were fly-overs, metro 
rails, atomic and thermal power plants, road projects and the rebuilding of infrastructure 
in Gujarat after the destructive earthquake in January 2001. In 2002 concrete 
consumption in India was estimated at 190 million m (249 million yd ). A. K. Jain 



45 



estimated that the rate of consumption of concrete should increase by at least 19.5 million 
m (26 million yd~ ) per year and ready mixed concrete should account for 11.0 million m 
(14 million yd ) of the total concrete produced in India by 2012 (as cited in Kumar & 
Kaushik, 2003). However, in 2008 the Indian Ready Mixed Concrete Manufacturer's 
Association (RMCMA) estimated that 20 to 25 million m 3 (26 to 33 million yd 3 ) of 
concrete were produced annually among 400 to 500 ready mixed facilities (RMCMA, 
2008). Although the ready mixed concrete industry in India is growing, the ready mixed 
concrete business in India is still emerging and only accounts for 5% of concrete 
consumed in India while the rest of concrete is site-mixed. 
Table 3.3 World Cement Production 2010 



Rank 


Country/Region 


Million Tonnes 


1 


China 


1880 


2 


India 


210 


3 


United States 


67.2 


4 


Turkey 


62.7 


5 


Brazil 


59.1 


6 


Japan 


51.5 


7 


Russia 


50.4 


8 


Iran 


50 


9 


Vietnam 


50 


10 


Egypt 


48 


11 


South Korea 


47.2 


12 


Saudi Arabia 


42.3 


13 


Thailand 


36.5 


14 


Italy 


36.3 


15 


Mexico 


34.5 


16 


Pakistan 


30 


17 


Germany 


29.9 


18 


Spain 


23.5 


19 


Indonesia 


22 


20 


Others 


480 


Total 


World 


3310 


(USGS, 2012) 





46 



1600 

1400 

•2 <% 1200 

-o I 1000 

Oh g 800 

IS 600 

u 400 

200 






■ China 

■ India 

□ U.S. 

□ Japan 



\~> 



n? 



sP 



$ 



# # # ^ 

& <& <£> 

f f f 



Figure 3.2 Trend in Cement Production for Four Leading Cement Producing 
Countries (USGS, 2012; Parikh, Sharma, Kumasr, Vimal, IRADe, 2009) 



1.40 



1.20 



73 0.40 



0.00 




■ China 

■ India 

□ U.S. 

□ Japan 



^ c> 



sf osr ^s ^ ' cn\ v 



$° oQ° J* 5 «© 



e .^ 



v v v A y> op vp 
& <& <& 

y* <Y r\? 



sy 



Figure 3.3 Potential Trend in Per Capita Cement Consumption for Four Leading 
Cement Producing Countries (USGS, 2012; Parikh, et al, 2009; United Nations 
2010b) 



47 



Table 3.4 Indian Cement Industry Informatio n 

India Cement Industry (Large Plants) 

Cement Companies 47 Nos. 

Cement Plants 142 Nos. 

Installed Capacity 222.61 million tonnes 

Production 160.75 million tonnes 

Domestic Despatches 158.25 million tonnes 

Per Capita Consumption 0.136 tonnes/person 
(CMA 2010a; CMA 2010c) 

3.3.1 Ready Mixed Concrete Industry in India 

The first ready mixed concrete batching plant was established in the city of Pune, 
India in 1987. The batching plant closed shortly after being unable to meet the demands 
of large projects like the Tanji Wadi subway, which led to skepticism in the RMC market 
in India (Alimchandani, 2007; Gordon & Kshemendranath, 1999). The first successful 
ready-mix concrete plant was ultimately set up 7 years after the Pune plant by ACC Ltd. 
(cement company) in Mumbai. Unitech Ltd. and RMC Group Pic soon followed with 
more ready mixed plants in Mumbai. However, even these RMC plants faced barriers 
because machinery and operations were not as sophisticated in comparison to plants 
established in Europe or South East Asia, agencies that could provide maintenance and 
technical support to plants were not well established in India, poor quality of aggregates 
led to inconsistent mixtures, the construction industry and contractors did not know how 
to schedule or plan for the use of ready mixed concrete, lack of specifications meant 
specifiers were reluctant to recommend a product they were unfamiliar with, and there 
was need to invest in training a workforce (Gordon & Kshemendranath, 1999). Since 
1994, the RMC industry has grown modestly. The perception of the RMC remains 
inconsistent since only a few companies have made the ready mixed industry their core 



48 



business while others such as cement companies, who own RMC companies, think of it 
as "downstream activity" for the main business of cement manufacturing (Gordon & 
Kshemendranath). Today, most RMC plants are located in major cities such as Delhi, 
Ahmedabad, Mumbai, Bangalore, Chennai, Kolkata, and Hyderabad where RMC 
accounts for 30 to 60% of total concrete used in the cities. To help encourage the growth 
of the RMC industry the RMCMA has been working towards guaranteeing quality RMC 
products through certification of plants around the country. 
3.3.2 Site Mixed Concrete in India 

The advancement of the construction industry in India has been slow due to many 
factors. Historically, construction in India was heavily dependent on government funding 
for infrastructure before India's government began encouraging private investment into 
developing infrastructure in 1991. As a result of construction projects being subsidized 
by the government, timelines for completion of the projects were not enforced, 
bureaucratic processes caused delays for construction, and the quality of construction 
projects was affected. Today, outdated construction techniques and specifications are 
still used. Site-mixed concrete still uses portable concrete mixers, human chains and 
wheelbarrows to transport the concrete, concrete buckets are lifted by mechanical winch, 
and steel rods are still being used for consolidation and compaction of concrete instead of 
vibrators. Many RMC companies encourage the use of RMC over site-mixed concrete 
and often list the following disadvantages of site-mixed concrete (Gordon & 
Kshemendranath, 1999; Lafarge, 2012): 

• The consistency and reliability of mixtures is dependent on the frequency of 

sampling and testing the variability of each mixture which is also dependent on 



49 



the variability occurring with the manual mixing of individual proportions of 50 
kg (1 10 lb) bags of cement 

• The volume of concrete production within an 8 hour shift is dependent on the 
skills of the laborer. 

• Manual mixing is more time consuming 

• The quality of raw materials is manually checked or not checked at all 

• Raw materials are often wasted 

• More money is spent on time, effort and laborers 

• Untrained and unskilled laborers create dangerous conditions and there is a lack 
of proper supervision 

• Since materials are stored on-site there is the likelihood that stock of materials can 
be stolen. 

Although, RMC companies make reasonable claims against site-mixed concrete RMC is 
still 12 to 20% costlier than site-mixed concrete. Additionally site-mixed concrete is still 
the dominant method used in construction in India especially for rural and developing 
urban areas as was seen with the city of Rajkot. Site mixed concrete is a major source for 
employment opportunities. The inclusion of site-mixed concrete in construction 
contributed to India's construction industry being recorded as the largest employment 
sector in 2000, thus employing 16% of the work- force available in India. This is 
significant in comparison to a 6 to 8% employment of the working population in 
developed countries (The Indian Concrete Journal, 2004). 



50 



3.3.3 Indian Concrete Mixture Designs 

Specifications up until the year 2000 were still based on 1950s construction techniques 
and old British Standards such that structural concrete was based on Ml 5 and M20 
grades of concrete. The batching occurred by volume, thus meeting a certain nominal 
ratio by volume for each grade of concrete. This meant minimum strengths had to 
achieve 15 MPa (2176 psi) and 20 MPa (2901 psi) respectively (Kumar & Kaushik, 
2003; Gordon & Kshemendranath, 1999). Today, these same mixture designs are 
commonly used for structural purposes in rural and developing urban areas. But as RMC 
concrete becomes more mainstream, specifications are revised to include more leeway for 
design mixed concrete, the roles of aggregate properties are better understood, benefits 
are seen with lower water cement ratios, and with research and development showing 
improved concrete strength from lower cement contents new grades of concrete (M20 
through M40) have been adopted by public works departments. Large construction 
projects have been known to use M50 grade which is a form of high strength concrete, 
high performance concrete, compacted reinforced concrete, reactive power concrete, and 
self compacting concrete. High grades of concrete are often used in bridges, piles, high 
rises, and power plants (Kumar & Kaushik, 2003). The use of waste materials or 
byproducts has increased (i.e. ground granulated blast furnace slag, metakaolin, and fly 
ash). Chapter 5 is dedicated to a discussion on fly ash use in Indian concrete. Table 3.5 
lists example mixture quantities for common grades of concrete in India. 



51 



Table 3.5 Mixture Proportions for Typical Grades of Concrete (Based on Saturated 
Surface Dry Conditions) 



Material 


M15 


M20 


M25 


M30 


M35 


M40 


3 

Cement kg/m 


270 


290 


320 


380 


400 


400 


3 

Water kg/m 


135 


145 


138 


160 


160 


160 


3 

Fine Aggregate kg/m 


711 


696 


751 


711 


704 


660 


3 

Coarse Aggregate kg/m 


1460 


1429 


1356 


1283 


1271 


1168 


20 mm kg/m' 


1051 


1029 


977 


924 


915 


701 


10 mm kg/m: 


409 


400 


380 


359 


356 


467 


Admixture kg/m 








1.6 


1.9 


2 


2.4 


water cement ratio 


0.5 


0.5 


0.43 


0.42 


0.4 


0.4 



(Kishore, 2012) 

3 3 

Conversion to U.S. Customary units is 1 kg/m' = 1.686 lb/yd 

3.4 Cement Manufacturing Process in India 

The manufacturing of cement in India is similar to the process described in a report 
published by the Portland Cement Association which focused on the life cycle inventory 
of portland cement manufacturing in the United States (U.S.) (Marceau, Nisbet , 
&VanGeem, 2010). The process used both in India and the U.S. is described in four 
major steps. Figure 3.4 shows the four steps in cement manufacturing, previously 
described, for a cement company in India known as Grasim Industries Limited. 

1 . Limestone quarries located near the cement plants are mined, drilled, and 
blasted to extract limestone. The limestone is crushed to approximately 5 cm (2 
in) and stored for blending. 

2. The limestone is proportioned with corrective raw materials in order to achieve 
the correct chemical composition. The materials are ground into a raw meal and 
stored in silos. Additionally any materials used for fuel (coal, wastes, petcoke 
and other alternative fuels) are processed, dried, and sized, blended and stored in 
silos onsite as well. 

52 



Limestone Mines and 
Crushing Plant 



Limestone and 
Coal Stockpiles 




Pre- 
Heater 



Clinker 
Loading 



Clinker 
Storage 



Cement Cement Mill Cooler 
Loading 



Raw 

Material 

Grinding 




Figure 3.4 Steps in cement manufacturing process at Grasim Industries Limited Cement Company (Grasim Industries 



w, Limited, 2008) 



3. The raw meal is fed into preheaters and then into the kiln systems. The fuel is 
fed into the kiln for combustion. High temperatures in the kiln help remove 
water from the raw meal, calcine the limestone, and cause necessary chemical 
reactions to form clinker. The clinker is cooled and stored before grinding. In the 
PC A report this stage is known as pyroprocess. 

4. The clinker is moved from storage. It is ground to a fine powder with gypsum 
and performance enhancer to make Ordinary Portland Cement (OPC). Fly ash or 
slag can be added at this stage to make Portland Pozzolana Cement (PPC) and 
Slag Cement, respectively. Cement leaves the plant in 50 kg bags or in bulk. 

3.4.1 Phases of Cement Clinker 

The process of making portland cement involves firing calcareous material (i.e. 
limestone, chalk, marl, and aragonite) with siliceous, argillaceous, and ferriferous ore 
materials (sand, shale, clay, and iron ore). The selection of raw materials is a meticulous 
process because high concentrations of trace elements can cause problems in the plant or 
in the final product. There are four main phases (Alit, Belie, tricalcium aluminate alkali 
solid solution, and ferrite phase solid solution) in the OPC that form once raw materials 
have reacted. Ideally the chemical compositions that represent these four phases are 
tricalcium silicate (3CaO.SiOi), dicalcium silicate (2CaO.Si02), tricalcium aluminate 
(3CaO.Al203), and calcium alumino ferrite (4CaO.Al203.Fe203). These chemical 
compositions are often abbreviated as C3S, C2S, C3A, C4AF such that C = CaO, S = Si02, 
A = AI2O3, and F = Fe203 (Gani, 1997). A phase diagram (refer to Figure 3.5) is best 
used to show how the relative proportions of the raw materials can direct the outcomes of 



54 



the phases and micro structure of the clinker. In general, OPC should fall within a C3S, 
C2S, and C3A triangle in a phase diagram (Gani, 1997). 



(m -kaolin 
AS2 from clay) 




Portland 
Cement 



from lime) 



mole% C 



Figure 3.5 Phase Diagram for Ordinary Portland Cement (Gani, 1997) 
3.4.2 Kilns 

The kiln plays an important role in contributing to the structure of the clinker and 
forming the final product. High temperatures are required to form the complex mixture 
of the clinker. The flame of the burner is approximately 2000°C (3632°F), the material 
making up the clinker has minimum temperature of 1455°C (2610°F), and precalciners 
are between 1000°C (1832°F) and 1200°C (2192°F) (WBCSD, 2005b; Gani, 1997). The 
kiln is usually a large steel tube lined with refractory (i.e. bricks) and is inclined by about 
3° to 5° from horizontal. The kiln rotates slowly (20 to 86 rph) as the raw materials are 
fed into the top of the kiln. There are three main types of processes used in the 
production of cement with a rotary kiln: wet, semi-dry, and dry process. In India between 



55 



2009 and 2010, 97.9% of cement produced by large plants was a result of using the dry 
process, 0.5% of cement production was completed by the wet process, and 1.6% of 
cement production was a result of other processes (CMA, 2010a). Within the wet process 
the raw materials are fed into the kiln as slurry with 37%-39% moisture due to being 
mixed with water. In the semi-dry process the raw material has 10%- 15% moisture and 
is partially calcined before entering the kiln. In the dry process the raw material is fed 
into the kiln as a dry powder. Cyclone heat exchangers and precalciners located before 
the kiln use the hot gases from the kiln to dry and partially calcine the raw materials. If 
precalcined, in addition to dried and preheated, the production rate in the cement kiln can 
be increased by 50% to 70%. To accomplish precalcining a burner is constructed 
between the kiln and the preheating cyclones. Precalcining can help extend the life of the 
refractory by reducing some of the heating load that is required in the kiln (Gani, 1997). 
3.5 Energy Consumption within the Cement Industry 

The production of cement is an energy intensive process. Particularly in step 3, of 
the cement manufacturing process, it was noted that high temperatures are required in the 
kiln. The traditional kiln fuels burned (coal, petroleum coke, sometimes natural gas, and 
fuel oil) result in an energy consumption between 3000 and 6500 MJ of fuel/tonne of 
clinker (depending on the manufacturing process) (WBCSD, 2005b). Grinding and 
milling are typically dependent on electricity and the pyroprocess might use electricity. 
Purchased electricity consumption can amount to 0.52 million Btu/tonne of cement (153 
kWh/ton of cement) (U.S. DOE, 2003). However, the global cement industry has the 
opportunity to increase efficiency by 0.2% to 0.5% per year, by replacing outdated 



56 



equipment, converting to the dry process, and focusing on mineral and energy recovery 
through use of wastes and by-products (WBCSD, 2005b). 
3.5.1 Energy Scenario in the Indian Cement Industry 

Indian industries such as steel, aluminum, and cement account for the largest 
share in the demand for commercial energy. In 2007 industries had a 44.8% share in the 
total energy consumption for India. The industry share could be further broken down 
into cement accounting for 13.5%, aluminum 11.4%, steel 39.7%, and others 35.4% 
(Dutta & Mukherjee, 2010). The Indian cement industry is the second largest producer of 
cement after China and has achieved world class efficiency following Japan's cement 
industry. Average kiln capacity is 2860 tonnes per day (3152 ton per day) which is 510 
tonnes (562 tons) less than Japan's kiln capacity (CMA, 2010a). The Indian cement 
industry has made significant modifications to the process in order to reduce the energy 
intensity. Technological upgrades have resulted in an average thermal energy 
consumption of 725 kCal/kg of clinker (2.6 million Btu/ton) and an average electricity 
consumption of 82 kWh/tonne of cement (0.3 million Btu/tonne) which is about 75 
kCal/kg of clinker (0.3 million Btu/ton) and 17 kWh/tonne (0.05 million Btu/ton) of 
cement more than that recorded for the best performing plant in the world (DRPSCC, 
201 1; CMA 2010a). In Table 3.6 energy use between India and the U.S. is compared. 

Between 2009 and 2010 Table 3.6 shows that the U.S. cement industries operated 
with lower energy efficiency than Indian cement industries. Additionally, India produced 
more clinker and cement while achieving lower energy intensities in that same year. As 
mentioned previously, India has invested in operational efficiency, process control, and 
energy conservation by use of alternative raw materials and fuels, waste heat recovery 



57 



systems/cogeneration systems, captive power plants, and higher productions of blended 

cement. 

Table 3.6 Average Energy Use Between India and U.S. Cement Industry for 2009- 
2010 



Energy Source or Material 


Unit 


India 


U.S. 




Fuel Energy Intensity 
Electricity Intensity 


GJ/tonne clinker 

(million Btu/ton) 

kWh/tonne of cement 

(million Btu/ton) 


3.0 (2.6) 
82 (0.3) 


4.2 
144 


(3.6) 
(0.4) 


Total clinker production 
Total cement production 


million tonne (million ton) 
tonne (million ton) 


128.3 (141.3) 
160.7 (177.1) 


56.1 
61.0 


(61.8) 
(67.2) 


Total Fuel Energy 
Total Electricity 


million GJ (million Btu) 
million kWh (million Btu) 


487.6 (462.1) 
13180.7 (45.0) 


255.0 

8784.0 


(241.7) 
(30.0) 



Source India: (DRPSCC, 2011; CMA, 2010a) 

Source U.S.: (Marceau, Nisbet, & VanGeem, 2010; USGS, 2011) 

3.5.2 Methods of Energy Efficiency 

As seen in Table 3.6, the electricity used (per tonne of cement) by Indian cement 
industries was about 57% of what the U.S. used. In order to avoid purchased electricity, 
Indian cement industries have established captive power plants (CPPs) on-site, where 
cement manufacturing occurs. Within 2002 and 2004 the installed capacity of captive 
power plants was growing faster than the country's generation utilities (Shukla, Biswas, 
Nag, Yajnik, Heller, & Victor, 2004). A common reason for the growth in CPPs was the 
advantage of having uninterrupted power for industrial processes. Unlike many of the 
power generation utilities for the country the CPPs are owned by the industries and not 
the government. However, in states such as Gujarat, permission to set up a CPP has to be 
attained from the Gujarat Electricity Board. The size of the CPPs can vary, for example, 
in the state of Gujarat, out of 163 CPPs in 2002, the smallest plant's installed capacity 
was 0.088 MW and the largest was 240 MW. The fuels that are commonly used in a CPP 
include lignite, coal, fuel oil, light diesel oil, high speed diesel, naptha, natural gas, and 
bagasse (fibers left from sugarcane). Cement industries are typical consumers of coal, 

58 



gas, and naptha and the typical sizes of the CPPs are medium (30 MW capacity) to large 
(above 50 MW capacity) (Shukla et al„ 2004; Ambuja Cements Ltd, 2010). Many 
industries that use CPPs use the plants as backups, but within the cement industry the 
CPPs provide the main advantage of a reduced cost in generation compared to tariffs 
established for industries by state utilities (Shukla, et al., 2004). Between 2009 and 2010 
59% of cement production in India was achieved with captive power plants. 

Another method of reducing energy demand within the cement manufacturing 
process is to use the method of waste heat recovery. Waste heat recovery leads to a 
reduction in fuel consumption which in turn could reduce the size requirements for the 
equipment needed for the waste heat recovery system and reduce emissions from 
combustion of fuels. Waste heat recovery systems in cement plants utilize hot gases for 
electricity production (also known as co-generation) or it can be used for preheating the 
raw material. Most waste heat from dry process cement kilns are within a temperature 
range of 620-730°C (1 148-1346°F) which is considered a medium temperature range 
(230-650°C [450-1200°F]) for waste heat recovery (BCS Incorporated, 2008). 
Preheating is the most common form of waste heat recovery and is accomplished by 
absorbing the waste heat from kilns and transferring the heat to the raw meal through 6 to 
4-stage cyclones that are located before the kiln (Refer to Figure 3.4). 

The efficiency of power generation depends on the temperature of the waste heat. 
Thus traditional waste heat recovery technologies need medium to high temperatures to 
produce power. To power an electric generator from waste heat, this can involve heating 
boilers to generate steam that turns a turbine. For cement kilns other technologies 
besides the traditional waste heat to boilers are being explored. These technologies 



59 



include organic Rankine and Kalina cycles. These technologies are being considered 
because they work more efficiently even with low to medium gas exhaust temperatures. 
The organic Rankine cycle uses an organic fluid (i.e. silicon oil, propane, isobutene, etc.) 
instead of steam with a higher molecular mass (desired for compact designs) and high 
mass flow to turn a turbine which will generate electricity. The Kalina cycle is similar to 
the Rankine cycle except it involves the use of ammonia and water as the working fluid. 
The combined use of fluids is called a binary fluid. Binary fluids can achieve greater 
efficiency because the boiling points of ammonia and water are different, therefore 
concentrations can be varied to attain more specific temperatures. Also, standard steam 
turbine components can be used if ammonia and water are used in a waste heat recovery 
system because both molecular weights (ammonia = 17.03 and water = 18.01) similar to 
steam so standard steam turbine components can be used in the waste heat recovery 
system (Mirolli, 2005; BCS Incorporated, 2008). 



Cyclones using 
waste gases from 

ktln 




Cyclones 
using hot 
combustion 
gases 



Combustion 
chamber 



Figure 3.6 Cyclone Heat Exchangers and Precalciner (Gani, 1997) 



60 



Use of by products and/or waste as fuels reduces the cement industries' demand 
for virgin fossil fuels and can reduce the industry's CO2 emissions. Hazardous and non- 
hazardous materials are sources of energy and can be used for fuel in cement kilns. The 
practice of using waste and by-products from other industries to create a closed-loop for 
resource use is also known as waste co-processing. This practice has been common 
among cement manufacturing industries in some parts of the world for more than 20 
years and is considered a method for waste management (i.e. Norway) (WBCSD, 2005b). 
To encourage safe and sustainable use of waste materials the Cement Sustainability 
Initiative established by the World Business Council for Sustainable Development has 
developed a document that provides guidance on the selection of fuels and raw materials 
for the cement manufacturing process (WBCSD, 2005b). Types of alternative fuels are 
listed in Table 3.7. However, the selection process for using alternative fuels depends on 
certain parameters, besides health, safety, and environmental considerations, which must 
be evaluated. For example, the assessment should be based on chlorine, sulfur, and alkali 
content (these constituents can clog the kiln system), water content, heat value, and ash 
content (ash content affects the chemical composition of the clinker). Any by-product or 
waste material must be introduced at the correct point in the cement manufacturing 
process in order to avoid unwanted emissions or changes in the necessary chemical 
composition of the clinker (WBCSD, 2005b). 



61 



Table 3.7 Examples of Non-Hazard ous and Hazardous Alternative Fuels 
Alternative Fuels 

Meat, bone meal, animal fat 

Tires 

Plastics 

Paper/wood/cardboard 

Coal slurries/distillation residues 

Sludge (sewage, water purification) 

Oil shales 

Agriculture, organic waste 

Paint residue 

Packaging waste 

Waste oil, oiled water 

Solvents 



(WBCSD, 2005b) 

3.6 Management, Energy Efficiency Ventures, and Emission Trends for Indian 

Cement Companies 

Understanding how the companies are managed can also explain why the Indian 
cement industry consumes less energy and still be able to produce more cement per year 
in comparison to a country like the U.S. (where Table 3.6 shows the differences in 
cement production). Periodically, cement companies in India will restructure and 
consolidate. For example, Gujarat Ambuja Cements Ltd. has a 14% stake in ACC 
Limited, Grasim Industries Limited acquired controlling stake over UltraTech in 2004, 
then Grasim vested with Samruddhi Cement in 2010 and finally merged with UltraTech 
(UltraTech, 2012; Dutta & Mukherjee, 2010). Additionally, the Indian cement industry 
comprises of some overseas investors. Stakes in Indian cement companies have been 
acquired by multinational companies such as Lafarge (acquired TISCO's operation) and 
Holcim (entered with Gujarat Ambuja) (Dutta & Mukherjee, 2010). Advantages of 
merging and reorganization of cement companies in India have evolved into the 
following: opportunity for the company to be highly competitive, have access to new 



62 



markets, and pursue cost effective and energy efficient technologies (Dutta & Mukherjee, 

2010). 

3.6.1 Energy Efficiency and Embodied in Cement Manufacturing in India 

The Cement Manufacturer's Association (2010a) from 2006 through 2010 
indicated that companies collaborated with the State Pollution Control Board and GTZ 
German Technology Corporation on trials of using waste derived fuels. The results of 
such collaboration led to recommendations for recycling hazardous wastes such as tires, 
paint sludge, petroleum tar waste, and effluent treatment plant sludge in the cement kiln. 
Cement companies such as Ambuja Cements received awards (such as the 2010 Green 
Tech Gold Environment Excellence Award and the 2010 National Award for Excellence 
in Water Management Award) emphasizing the company's investment in energy efficient 
technologies. Ambuja has also indicated that 70% of total power requirement in 201 1 was 
generated from the captive power plants. An article in "The Hindu Business Line" 
indicated that a 1 million tonne cement plant would need about a 20 MW of power 
capacity and according to the Grasim Sustainability Report a combined capacity of 144 
MW captive power plants are located at four sites. So it might be assumed that on-site 
captive power plant capacity could range between 1 MW to 40 MW depending on the 
capacity of cement production (Ramakrishnan, 2012 & Grasim Industries Ltd., 2008). 
India Cements Company has an 8 MW waste heat recovery plant. ACC Ltd. Cement 
Company uses captive power to meet 72% of its power requirement (Ramakrishnan, 
2012). Grasim Industries began utilizing hazardous waste in kilns since 2007-2008 and 
reported that 1,400 tonnes of coal was replaced with 2,823 tonnes of hazardous waste 
between 2007 and 2008. Grasim has setup a municipal solid waste processing plant such 



63 



that the processed waste used as alternative fuel in 2007 was 7126 tonnes. Between 2007 
and 2008 Grasim received the Energy Conservation Award and the Greentech Silver 
Award for reductions in dust emissions. Grasim was one of the earliest users of the 
rankine cycle technology for waste heat recovery (Grasim Industries Limited, 2008). 
Within the annual reports prepared by the individual companies information about how 
the company conserved energy or upgraded equipment within plants is reported. 

Understanding technological upgrade and methods of generating energy and fuel 
use within the Indian cement industry was important for this study in order to verify or 
calculate a cement emission factor. As will be explained in the section pertaining to the 
life cycle analysis (LCA) of cement a cement emission factor had been calculated by a 
few organizations or entities within the country of India, however, these emission factors 
did not agree with one another. Therefore, as part of this study it became pertinent to 
perform a bottom-up approach to calculate or verify the Indian cement emission factor 
which required a little more in-depth knowledge about individual companies. Since, the 
case study involved the city of Rajkot the Indian cement companies that were located in 
Gujarat were used for the performance of the LCA. 

According to the CMA (2010c) between 2009 and 2010 there were at least eight 
different member companies that had plants in the state of Gujarat. Three of the member 
companies (Gujarat Sidhee Cement, Saurashtra Cement [known as the brand Hathi] and 
Ultratech Cement Ltd.) and two non-member companies (Grasim Industries and Ambuja 
Cements Ltd.) were chosen for the LCA study. 



64 



3.6.2 Emission Trends in Cement Manufacturing in India 

GHG emissions have been associated with the charge of contributing to climate 
change. As stated previously the importance of quantifying these emissions leads to 
comprehension of material use and embodied energy of these materials. Finally, methods 
for reducing emissions depend on revolutionizing the way materials are used and 
modifying the embodied energy associated with the materials. Total emissions can be 
calculated by multiplying an emission factor (EF) by the total amount of activity or 
production of a material. The EF or emission intensity is the rate of a pollutant or gas 
relative to the activity or production of material (IPCC, 1996). 

The CO2 emissions from the production of cement are a function of two 
processes: calcining and the combustion of fuel. Calcining is the process when the raw 
material chemically changes when reaching extremely hot temperatures. In other words 
when heating the calcium carbonate (CaCC>3), coming from calcium rich materials (i.e. 
limestone), calcium oxide (CaO) and carbon dioxide (CO2) form (see also Equation 3.1) 

CaC0 3 + Heat -> CaO + C0 2 (3.1) 

Estimation of CO2 emissions from calcining is a function of the lime (CaO) percentage 
(content) for clinker. In the IPCC 1996 guidelines the default lime content was 
estimated at 0.646. Lime percentages vary little between cement plants so if lime content 
is unknown the IPCC default factor is often used (WBCSD, 2005a). Lime content can 
result from other materials such as fly ash and not from the calcium carbonate. If that is 
the case this percentage of lime content should be subtracted out of the total lime content 
before calculating the calcining emission factor (IPCC, 2006). The lime content is 
multiplied by the molecular weight ratio for CCVCaO (44.01 g/mole -=- 56.08 g/mole = 



65 



0.785) to calculate tonnes of CCVtonne of clinker. Thus the emission factor (EF) for 
calcining, in reference to clinker produced, is 0.507 tonnes CCVtonne of clinker (IPCC, 
1996). In the 2006 IPCC guidelines a correction factor for cement kiln dust (CKD) was 
incorporated into the calcining emission factor. CO2 can result from lost CKD and can 
range between 1.5 and 20% for a cement plant. If no information is available on CKD 
the default factor recommended by IPCC (2006), is 1.02. The 0.507 tonnes CCVtonne of 
clinker factor is multiplied by the CKD correction factor (See Equation 3.2. The 
corrected calcining emission factor is 0.517 tonnes CCVtonne of clinker. 

EF C H n ker = lime content xmolecular weight of COi/CaO xCKD correction factor 

EF C H nker =0.646 x0.785 xl.02 
EF dinker = 0.517 tonnes COi/tonne of clinker (3.2) 

The general methodology for estimating emissions from the combustion of fuel 
and electricity used requires the knowledge of the total amount of fuel or energy used in 
the process, and the emission factor that relates the rate of CO2 released per amount of 
fuel combusted or electricity used. The amount of fuel used in the process can be 
reported as total volume, mass, or energy. Additionally, the emission factor can be 
reported as rate of CO2 released relative to energy associated with the fuel combusted. In 
these cases the calorific value (i.e. kcal/ kg) and density of the fuel (kg/m ) is needed in 
order to derive a final emissions factor in the form of tonnes of CO2 per tonne of cement 
produced. 

This study involves the calculation of an emission factor for cement, but the 
government of India, the Cement Manufacturers' Association, as well as a few individual 
cement companies have established cement emission factors. However, recent (years 



66 



2007-2010) emission factors, even if for the same year, vary between 0.65 - 0.83 tonne 
CCVtonne of cement material (Refer to Figure 3.7). From Figure 3.7 there is a 
downward trend in emissions starting from 1996 to 2010. The decrease in emissions is 
best explained by the upgrade of equipment for energy efficiency, the use of captive 
power, cogeneration, clinker substitution (with raw materials such as fly ash and slag), 
and wind power generation (CM A, 2010a). 



•- 

O 



1.2 



1 



u 

ta g 0.8 
§ S 

.a ^ 0.6 

c ^ 
W O 
?U 0.4 

r 

3 °- 2 



A 

1 



i4+ 



B 



o 4M-4 



1 



E,C 



H,G 






ll. M 



■ ■ 




T I I I I I I I I I I I I I I I I 

HMmi-ChoooNOH^fn^iri^hooao 

ONONONONONONONONOOOOOOOOOO^H 
ONONONONONONONONOOOOOOOOOOO 



Year 



Sources: A - Schumacher, Sathaye, 1999; B - Hendricks, Worell, de Jager, Blok, 
Riemer, 2004; C - Parikh, Sharma, Kumar, Vimal, IRADe, 2009; D - CCAP, TERI, 
2006; E - Garg, Shukla, Kaphse, 2006; F - MoEF, 2010; G - CMA, 2010a; H - 
WBCSD, 2010 
Figure 3.7 Indian Cement Emission Factors for 1991-2010 

Emission factors reported in 2006 and 2007 were a result of the CMA taking part 

in a two phase project, under the Ministry of Environment (MoEF) and Forests and 

United Nations Framework Convention on Climate Change (UNFCCC), titled NATCOM 

(National Communication). The project involved annualizing GHG emissions for 

developing countries who were participants in the Kyoto Protocol. The CMA gathered 

data to calculate emissions from 1 19 major plants out of 136 in 2007 (this was 



67 



approximately a 88% response to a questionnaire of questions sent out to the companies). 
During the process for gathering data for emissions, the CMA encountered certain 
problems. For example, many of the major companies (their capacity accounted for 40% 
of the total capacity of the country) were not CMA members and had chosen to report 
emissions using the system affiliated with the Cement Sustainability Initiative established 
by the World Business Council on Sustainable Development. These companies included 
Grasim Industries [now under UltraTech], ACC Ltd., and Ambuja Cements Ltd. In 
Figure 3.7 for the years 2007 through 2010, individual cement companies had 
volunteered to report company emissions through the Cement Sustainability Initiative 
(CSI), Both CMA and CSI follow IPCC guidelines; however, according to the CMA 
there are some differences between CMA and CSI in their process for emissions 
calculations which were not explained. Upon review of the guidelines by CSI the 
differences could lie in the emissions factors for the fuels. CSFs guidelines are meant for 
use by many companies around the world so only default emission factors listed in IPCC 
and a few CSI calculated emission factors are listed in the guidelines. It is possible the 
cement companies in India who are working under the CSI protocol may have not used 
country specific emission factors for fuel as has been done by the MoEF and CMA. A 
list of fuel and electricity emission factors is shown in the Appendix as Table B.l. The 
purpose of Table B.l is to show users how important it is to research the correct fuel 
emission factor because often a fuel may be called something different between countries 
but is essentially the same fuel or sometimes the fuel emission factor can be updated 
every few years. Additionally, in order to create Table B.l it involved an intensive 
literature review. At the time of the study there was not a well established database of 



68 



country specific emission factors. Current comprehensive online emission factor 
databases or up-to-date life cycle inventory data are not free and require the user to pay a 
certain fee. Such databases include http://www.ecoinvent.org/ and 
http://emissionfactors.com/ 

Calcining emission factors can differ and Table 3.8 shows some examples. CSI 
provides guidance to companies on how to calculate CO2 emissions from calcination of 
clinker, dust, and carbon from raw materials. The guidelines specifically encourage 
companies to measure calcium Oxide (CaO) and magnesium oxide MgO contents of 
clinker at the plant level (WBCSD, 2005a). These measurements can give a more precise 
emission factor for calcination and will most likely differ from IPCC's default 0.517 
tonnes CCVtonne of clinker. Grasim, in their 2007-2008 sustainability report, calculated 
a calcining emissions factor that differed from IPCC's default by about 20%. The CMA 
also calculated their own calcining emission factor (which includes a CKD factor) as 
0.537 tonnes CCVtonnes clinker produced. The values reported by GRASIM and 
MoEF/CMA could be representative of a range of India's calcining emission factors. For 
this study it was decided that CMA's calcining emission factor of 0.537 tonnes 
C02/tonne of clinker would be used in the life cycle analysis. 

Table 3.8 Example Differences in Calcining E mission Coefficients 

Calcining 
emissions 
Source Year (tonne 

C02/tonne of 
clinker) 



MoEF/CMA 2010 


0.537 


GRASIM/WBCSD 2008,2010 


0.555 


IPCC 1996, 2006 


0.517 



Note: Grasim actually reported a calcining emission factor of 0.427 tonnes C02/tonne of 
cement. The Grasim sustainability report, however, did not show total clinker produced, 

69 



thus a cement/clinker ratio was assumed based on the industry average for the year 
2007-2008 in order to convert to tonne C02/tonne of clinker. The industry average for 
cement/clinker ratio was 1.45 and for a company like UltraTech (who merged with 
Grasim) its ratio was 1.14. So the average between the two ratios (1.3) was used in 
order to calculate calcining emissions for Grasim (Source: 
http://content.icicidirect.com/mailimages/Ultra tech-final.pdf) 

3.7 Materials, Fuels, and Emissions Associated with Cement and Concrete 

3.7.1 Cement 

Based on CMA's cement emission factor and the emission factors reported by 
individual companies through the CSI it was uncertain which was most representative of 
the Indian cement industry. Observations of cement use in Rajkot led to the decision to 
determine a cement emission factor based on frequently used brands of cement in Rajkot. 
The brands included Ambuja Cement Ltd., UltraTech Cement Limited, Gujarat Sidhee 
Cement Limited, Hathi Cement (brand name under flagship company Saurashtra Cement 
Limited). Using the annual reports published through company websites, a year's worth 
of data was gathered either for 2009-2010 or 2010-201 1 regarding the electricity 
purchased, total energy used from coal, total volume of certain fuels and oils, total clinker 
produced, and total cement produced. Typical data gathered from the annual reports are 
shown in Figure B.l in Appendix B. The annual reports provided the opportunity to 
determine which companies were taking advantage of certain technologies (as discussed 
in Section 3.4.2) that made the manufacturing process more energy efficient. Table 3.9 
lists all the raw data gathered from the four companies. The clinker/cement ratio was 
calculated from the production of cement and clinker that was reported on the annual 
reports. From Table 3.9 it is important to note that all companies reduced the dependence 
on grid electricity through the use of captive power plants. Major companies such as 
UltraTech showed the use of waste heat recovery. 

70 



Table 3.9 Fuel and Electricity Raw Data Gathered for Calculation of Cement Emission Factor 



2009-2010, 2010-2011 Data 


Ambuja Cements 


Ultra Tech 


Sidhee Cements 


Ha tlii Cements 


Average 


Electricity Purchased (Kwh) 


402000000 


361072000 


108177000 


1079000 


218082000 


Total Rs. 


1696900000 


1812100000 


676616000 


15148000 


1050191000 


Rate (RS/Unit (Kwh)) 


4.22 


5.02 


6.25 


5.68 


5.29 


Electricity Generated (Kwh) 


186400000 


61264000 


167000 


1259000 


62272500 


Net Units/Ltr. Of Light Diesel Oil/Furnace oil 


3.9 


3.93 




3.13 


3.65 


LDO/furnace oil cost (Rs)AJnit Generated 


7.02 


6.99 




8.77 


7.59 


Fuel cost/electricity duty 






16.04 




16.04 


Electricity Steam Generator (Kwh) 


1209300000 


1187204000 




132248000 


842917333 


Net Units /T of Fuel 


842 


1030 






936 


Oil/Gas Cost/unit 


3.14 


3.17 




3.85 


3.39 


Total Amount (Rs) 








508910000 




Waste Heat Recovery System (kWh) 




13997000 






13997000 


Cost/Unit 




0.4 






0.4 


Coal (million KCal) 


10533678 


18410858.27 


937363 


1026672 


7727142.82 


Cost (Rs.) 


8930000000 


10861700000 


970970000 


1033487000 


5449039250 


Average Rate (Rs/million K. Cal) 


847.53 


589.96 


1035.85 


1006.64 


870.00 


Light Diesel Oil/High Speed Diesel (K. liters) 


3508.87 


1112.00 


184.79 




1601.89 


Cost (Rs.) 


126900000 


39900000 


7563000 




58121000 


Average Rate (Rs/K. liters) 


36178 


35903 


40926.35 




37669.12 


Furnace Oil (Including Naphtha) (K. liters) 




22692 




682 


11687 


Cost (Rs.) 




488500000 




16662000 


252581000 


Average Rate (Rs/K. liters) 




21527 




24431 


22979.04 


High Speed Diesel Oil (HSD) (K. liters) 




3154 






3154 


Cost (Rs.) 




110000000 






110000000 


Average Rate (Rs/K. liters) 




34861 






34861 


LDO (Liter)/Tonne of clinker 


0.24 


0.11 


0.16 




0.17 


Coal and other fuels (K. Cal/Kg. of Clinker) 


750 


709 


811 


802 


767.93 


Electricity (Kwh/Tonne of cement) 


85.9 


83.13 


86.21 


102.85 


89.52 


Total Cement Production (tonnes) 


20100000 


17639000 


1211754 


1158720 


10027368.5 


Clinker sold 


343525 


2461000 


29725 


202231 


759120.25 


Clinker Produced (tonnes) 


14100000 


15550000 


1160000 


1280610 


8022652.5 


Ratio (clinker/cement) 


0.70 


0.88 


0.96 


1.11 


0.9 1, 



The electricity and fuel data was converted to total CO2 emissions using country 
specific emission factors available from various sources listed in Table 3.10. As seen in 
Table 3.9 not all the fuel information was recorded in terms of energy. For the fuels that 
were recorded in units of volume, information such as calorific value of the fuel and 
density of the fuel were required. The calorific values are included in Table 3.10 and 
density values are shown in Table 3.11. An average density was calculated within each 
range shown in Table 3.11 and was used in the calculations for total CO2 from the fuel 
used. The equations below are shown to clarify the process used to determine the unit 
mass of CO2 from total fuel used in the cement manufacturing process for the year. Note: 
All fuel for on-site transportation was assumed to be included in the data provided in the 
annual reports. If transportation energy use was not reported as part of the annual 
reports then according to the study performed by Marceau, Nisbet, and VanGeem (2010) 
we can assume transportation energy contributes about 2% of total energy input. 
Marceau, Nisbet, and VanGeem calculated an average of 0.091 G J /tonne of cement 
(39.1Btu/lb cement) and 3.2 kgC0 2 /tonne of cement (6.41 lb ofC0 2 /ton of cement) from 
transportation. 

total mass of fuel = total liters of fuel x density 
total energy of fuel = total mass of fuel x color fie value 

unit mass of C0 2 



unit mass of C0 2 = total energy from fuel x 



unit energy of fuel 



or 



unit mass of C0 2 

unit mass of C0 2 = total mass of fuel x — — - 

unit mass of fuel 



72 



Table 3.10 Country Specific Emission Factors Used in Calculating a Cement 
Emission Factor 



Fuel/Electricity kg C0 2 /kWh Source and Additional Information 



Electricity Coal 
Purchased 



Coal 



Furnace oil 



Natural Gas 



0.83 



0.35 




0.28 




0.20 



CEA, 2009 

The value was the average of coking, non- 
coking, and lignite which is the usual 
Indian coal fuel mix. An average was 
taken from the following Indian EF: 
93.61, 95.81, 106.15 tonnes CO/TJ 
(MoEF, 2010) 

Both the NCV and EF were used to 
calculate the EF in terms of C0 2 per 
energy. The Indian EF = 3 . 1 8 tonnes 
CCytonne and NCV = 43.33 
TJ/kilotonnes (Ramachandra and 
Shwetmala, 2009) 

Furnace oil is also called fuel oil The 
Indian EF = 77.4 tonnes C0 2 /TJ 
Both the NCV and EF were used to 
calculate the EF in terms of C0 2 per 
energy. The Indian EF = 3. 1 8 tonnes 
C0 2 /tonne and NCV = 43.33 
TJ/kilotonnes (Ramachandra and 
Shwetmala, 2009) 
CCAP & TERI, 2006 EF actually 
reported as 55.82 tonnes of C0 2 /TJ 




1 kg C0 2 /kWh = 646 lb/MBtu 



Table 3.11 Density Values for Certain Fuels Used in Indian Cement Manufacturing 



Fuel/Oil 



Density (kg/m ) Source 




Light Diesel Oil 



Furnace Oil 



High Speed Diesel Oil 



820-880 



890-950 
820-860 



" 



1 kg/m" = 0.062 lb/in" 



Indian Oil Corporation Ltd 
(http://www.iocl.com/) 
Bureau of Energy Efficiency 
Fuels and Combustion 
Guidelines (www.em-ea.org) 
Indian Oil Corporation Ltd 
(http://www.iocl.com/) 



73 



Tables 3.12 through 3.19 demonstrate how the energy reported in the annual reports were 
converted into total emissions. Energy is considered the material flow analysis portion 
while the emissions factor for each type of fuel used for energy is conisdered the life 
cycle analysis results. Finally MFA multiplied by LCA results into total impact or total 
emissions from the energy produced from the fuel. 
Table 3.12 MFA-LCA Data for Purchased Electricity 



Purchas e d Ele ctricity | 


Company 


MFA 
(kWh) 


LCA 
(kgC0 2 /kWh) 


Total Emissions 

(kgC0 2) 


Ambuja 


" 4.02E+08 


0.83 


3.E+08 


UltraTech 


r 3.61E+08 


0.83 


3.E+08 


Sidhee 


r 1.08E+08 


0.83 


9.E+07 


Hathi 


T 1.08E+06 


0.83 


9.E+05 



Table 3.13 MFA-LCA Data for Company Generated Electricity from Coal 



Own Ge ne ration Ele ctricity (Coal) j 


Company 


MFA 
(kWh) 


LCA 
(kgC0 2 /kWh) 


Total Emissions 
(kgC0 2 ) 


Ambuja 


r 0.00E+00 




0.00E+00 


UltraTech 


" 1.19E+09 


0.60 


7.16E+08 


Sidhee 


r 0.00E+00 


0.60 


0.00E+00 


Hathi 


" 1.32E+08 


0.60 


7.97E+07 



Table 3.14 MFA-LCA Data for Company Generated Electricity from LDO/Furnace 
Oil 



Own Generation Electricity (LDO/Furnace Oil) | 


Company 


MFA 
(kWh) 


LCA 
(kgC0 2 /kWh) 


Total Emissions 
(kgC0 2 ) 


Ambuja 


1.86E+08 


0.46 


8.50E+07 


UltraTech 


' 0.00E+00 


0.46 


0.00E+00 


Sidhee 


" 1.67E+05 


0.46 


7.62E+04 


Hathi 


' 1.26E+06 


0.46 


5.74E+05 



74 



Table 3.15 MFA-LCA Data for Company Generated Electricity from Natural Gas 



Own Generation Electricity (Natural Gas) 


MFA LCA 
Company 

v J (kWh) (kgC0 2 /kWh) 


Total Emissions 
(kgC0 2 ) 






UltraTech 6.13E+07 0.32 


2.0E+07 


Sidhee ' 0.00E+00 0.32 


0.0E+00 


Hathi r 0.00E+00 0.32 


0.0E+00 



Table 3.16 MFA-LCA Data for Thermal Energy from Coal 



Thermal Energy (Coal) 


Company 


MFA 
(kWh) 


LCA 
(kgC0 2 /kWh) 


Total Emissions 
(kgCQ 2 ) 


Ambuja 


' 1.22E+10 


fltvj 


4.3E+09 


UltraTech 


' 1.34E+10 


0.35 


4.8E+09 


Sidhee 


" 1.09E+09 


0.35 


3.9E+08 


Hathi 


" 1.19E+09 


0.35 


4.2E+08 



Table 3.17 MFA-LCA Data for Thermal Energy from Light Diesel 



Thermal Energy (Light Diesel) 


Company 


MFA 
(kWh) 


LCA 
(kgCQ 2 /kWh) 


Total Emissions 
(kgC0 2 ) 


Ambuja 


" 3.71E+07 


0.35 


1.3E+07 


UltraTech 


" 1.18E+07 


0.35 


4.2E+06 


Sidhee 


r 1.95E+06 


0.35 


6.9E+05 


Hathi 


" 0.00E+00 


0.35 


0.0E+00 



Table 3.18 MFA-LCA Data for Thermal Energy from Furnace Oil 



Thermal Energy (Furnace Oil) 


Company 


MFA 
(kWh) 


LCA 
(kgC0 2 /kWh) 


Total Emissions 
(kgC0 2 ) 


Ambuja 


" 0.00E+00 


0.28 


0.0E+00 


UltraTech 


" 2.55E+08 


0.28 


7.1E+07 


Sidhee 


r 0.00E+00 


0.28 


0.0E+00 


Hathi 


" 7.66E+06 


0.28 


2.1E+06 



75 



Table 3.19 MFA-LCA Data for Thermal Energy from High Speed Diesel Oil 



Company 



Thermal Energy (High Speed Diesel Oil) 



MFA LCA Total Emissions 

(kWh) (kgC0 2 /kWh) (kgC0 2 ) 







n ?fi 






UltraTech 


' 3.19E+07 


0.26 


8.4E+06 




Sidhee 


' 0.00E+00 


0.26 


0.0E+00 


■ 


Hathi 


" 0.00E+00 


0.26 


0.0E+00 


.1 



Table 3.20 shows the total production of cement for a given year for each company. 

Table 3.20 Cement Production from Major Cement Manufacturing Companies that 
Deliver to Rajkot, India 



Cement Production Ambuja UltraTech Sidhee Hathi 



to: 



Cement (million 
tonnes cement) 




20.1 



3.7.1.1 Overall Result 

Table 3.21 and Table 3.22 show the results of energy use and CO2 emissions 
arising from the fuels and electricity per unit production of cement. The calculations were 
based on four major cement companies that have plants located in Gujarat and provide 
cement to city projects based in Rajkot, India. From Tables 3.9 and 3.22, the data 
revealed that Ambuja and UltraTech are the larger producers of cement and total CO2 
emissions. All companies do use captive power plants to save on costs spent on 
purchased electricity either by generating electricity through fuel oils and coal (Refer to 
Table 3.9). Companies such as Ambuja and UltraTech appear to be using natural gas as 
well (Ambuja, 2010; UltraTech, 2011; and Shukla et al., 2004). UltraTech in particular 
reported some energy savings through the use of waste heat recovery. The savings 
totaled about 0.002GJ/tonne of cement (1.2 Btu/lb of cement). Finally, averaging the 
four main Gujarat cement producing companies revealed that the average cement 
emission factor is approximately 0.84 tonnes C02/tonne of cement (1680 lb COi/short 



76 



ton of cement) when excluding purchased electricity emissions. This average is closer to 

the emission factor reported by CMA (2010a) for the year 2007. 

Table 3.21 Energy Consumption from Major Cement Manufacturing Companies 
that Deliver to Rajkot, India. 



Ambuia UltraTech Sidhee Hathi 



Thermal 
(kWh/tonne cement) 



610.92 776.61 900.66 1036.38 






Own Generation 

Electricity 69.44 70.78 0.14 115.22 

(kWh/tonne cement) 



Purchased Electricity 
(kWh/tonne cement) 



20.00 20.47 89.27 




Total Energy 
(kWh/tonne cement) 



700.36 867.86 990.07 1152.53 



* 



Net purchased electricity 
1 GJ/Tonne of cement = 429.92 Btu/lb of cement, 1 kWh/Tonne of cement = 1.54 Btu/lb, 
1 tonne/tonne = 2000 lb/short ton = 1000 kg/tonne 

3.7.1.2 Company to Company Comparison 

From Table 3.21, key cement producers such as Ambuja might be expected to 

produce more CO2 per tonne of cement since the industry uses more energy from both 

thermal and purchased electricity sources at least compared to Sidhee and Hathi. 

However, once the energy is calculated per unit mass of cement and converted to CO2 per 

tonne of cement, a large cement producer such as Ambuja demonstrates that it has taken 

certain measures to reduce energy consumption and CO2 emissions for the large amounts 

of cement that they produce. In the annual reports, reduction in CO2 is not discussed, 

however, details on energy conservation are required to be listed as per Section 217 (1) 

(e) of the Companies Act, 1956 (Ambuja Cements Ltd., 2010). Energy saving methods 

77 



were discussed earlier in this chapter and Ambuja uses some of these methods as listed in 
their annual report. These methods include optimization and upgrading the process and 
equipment (replacing pre-heaters, shortening the chamber for the cement mill), adjusting 
operating voltage for lights, installing alternate fuel system lines, and installation of more 
captive power plants. Although, Ambuja is using waste derived fuels and has an 
alternative fuels and raw material (AFR) testing laboratory, the amount of waste used as 
fuel is not reported in the annual reports. In fact, the other three companies' reports did 
not include the amount of alternative waste fuel used. Any additional CO2 coming from 
waste fuels is not being included in the calculation of the cement emissions factor so the 
emissions factor might be underestimated in this dissertation. However, we might 
assume that the alternative waste fuels may not contribute more than say 5% of total CO2. 
Fossil fuels are still dominant in the entire cement process and this could be a valid 
assumption because, if we recall that Grasim seemed to be the first to provide a thorough 
emissions and energy inventory. Within Grasim's inventory it was calculated that 
alternative waste fuels contributed about 0.6% to the total CO2 emissions reported from 
kiln fuels for Grasim. It is also important to note that Grasim, ACC Ltd., and Ambuja 
were the companies that had the largest share in the industrys' capacity for the year 2010. 
Therefore, it should be safe to assume that all other companies' emissions from 
alternative waste fuels were 0.6% or less of total CO2 emissions. This percentage should 
not greatly change the calculations presented in Table 3.22. 



78 



Table 3.22 Emissions from Major Cement Manufacturing Companies that Deliver 
to Rajkot, India. 



Emissions Ambuja UltraTech Sidhee Hathi 



Thermal 

216.68 274.19 319.45 367.09 



(kgC0 2 /tonne cement) 



Own Generation 

Electricity 23.69 41.71 0.06 69.31 

(kgC0 2 /tonne cement) 









Purchased Electricity 
(kgC0 2 /tonne cement) 



16.60 16.99 




Calcining 

_ ___ . " N 376.70 473.40 514.06 593.49 

(kgC0 2 /tonne cement) 



. 



Total Emissions 



633.68 806.29 907.67 1030.66 

(kgC0 2 /tonne cement) 

1 kg/tonne = 0.001 tonne/tonne, 1 tonne/tonne = 2000 lb/short ton = 1000 kg/tonne 
3.7.1.3 Cementitious Materials 

Additionally, Ambuja, as well as the other three companies, use fly ash and slag 
to produce blended cements and thus reduce the amount of clinker that is needed in the 
process, which leads to fewer C0 2 emissions. Not all companies reported the amount of 
slag used for the year so in Table 3.23 only the amount of fly ash used for the year 2009- 
2010 or 2010-201 1 is shown. Ambuja uses fly ash equal to about a quarter of how much 
cement is produced, UltraTech uses an amount of fly ash that is at least 15% of the 
cement produced and the other two companies use of fly ash between 4 and 5% of the 
amount of cement they produce. 



79 



Table 3.23 Fly Ash Consumption by Major Cement Companies who Deliver to 
Rajkot, India 



Fly Ash Consumption Ambuia Cements Ultra Tech Sidhee Cements Hathi Cements 



Fly ash (million tonnes) 


4.97 


2.59 


0.05 


0.06 


fly ash/cement ratio 


0.25 


0.15 


0.04 


0.05 


3.7.1.4 Energy Intensity 











Recall in Table 3.6 the average energy intensity for cement manufacturing in 
India according to the Cement Manufacturers' Association (2010a) and DRPSCC (201 1) 
was shown. The fuel energy intensity according to the CMA is including only fuels used 
for firing the kiln while the electricity intensity is including the purchased and onsite 
generation of electricity. The annual reports did not make it clear whether the "own 
generation" (or onsite electricity) was included as the fuels listed. But after back 
calculating the fuels needed for on-site electricity the total amount of fuel did not match 
the amount of fuels reported in the annual reports. So the energy reported for on-site 
generation was converted into CO2 emissions using each of the fuel emission factors 
reported in Table 3.9 and included efficiency for coal, oil, and natural gas captive power 
plants. The efficiency of the captive power plants were taken from a report written by 
CCAP and TERI (2006) where coal was 30% efficient, oil was 32% efficient, and natural 
gas was 39% efficient. Thermal energy reported in Table 3.21 can be compared to Table 
3.6. If the energy in Table 3.21 is converted using the clinker cement ratio reported in 
Table 3.9 the values would change such that Ambuja = 3.1 GJ/tonne of clinker (1344 
Btu/lb of clinker), UltraTech = 3.2 GJ/tonne of clinker (1359 Btu/lb of clinker), Sidhee = 
3.4 GJ/tonne of clinker (1444 Btu/lb of clinker) and Hathi = 3.3 GJ/tonne of clinker 
(1438 Btu/lb of clinker). All cement companies report a slightly higher thermal energy 
intensity compared to the average reported in Table 3.6 for India. Electricity intensity 

80 



was reported in the annual reports (See Table 3.9), however, trying to recalculate this 
same intensity using the values available from the report resulted in different electricity 
inensities as can be seen in Table 3.21. It is possible that the electricity reported in the 
annual reports included electricity used for colonies (or the people and villages that live 
on-site or next to the cement manufacturing plants). Nevertheless, using the electricity 
intensities reported in Table 3.21 (Own + Purchased) it can be seen that all cement 
companies use more electricity per tonne of cement compared to the average reported in 
Table 3.6, with the exception that Amubja and Sidhee electricity intensities being only 7 
kwh/tonne of cement (10.8 Btu/lb of cement) more than the average. 
3.7.1.5 CO g Emissions Factor Conclusion 

From Table 3.22 companies, such as Sidhee and Hathi, whose production is only 
8% to 9% of the production of Ambuja produce a maximum of 63% more tonnes of 
CCVtonne of cement compared to Ambuja. It is worth noting that Sidhee appears to 
still be heavily reliant on purchased electricity compared to Hathi. Although Hathi 
purchases less electricity, the resulting total tonne of CO2 emissions per tonne of cement 
is about 13% higher than Sidhee assuming all cement manufacturing companies share the 
same calcining emissions. UltraTech and Ambuja show emission factors lower than 
Grasim (refer back to Table 3.2). This confirms that Ambuja, UltraTech, and Grasim are 
still leaders in investing in energy efficient and CO2 reducing methods. Although all 
calcining emissions were assumed to be based on CMA's (2010a) calculations, this 
dissertation proves that it was necessary to go through the process of calculating a cement 
emission factor for at least Gujarat, India and clarify where the cement manufacturing 



81 



process stands in India in terms of CO2 emissions as compared to the various emission 

factors shown in Figure 3.7. 

3.7.2 Quarrying and Mining of Other Raw Materials (Excluding Limestone) 

Limestone is quarried onsite where cement is manufactured. However, its not 
clear whether raw materials (sand, gypsum, clay) included in the cement process are 
mined on-site or off-site. Table 3.24 shows the amount of material mined or quarried and 
the total emissions from the equipment used in the process. Off-site information was 
used to determine an emission factor for quarrying and mining these raw materials. 
Using GRASIM's information, cement manufacturing uses approximately 0.45 tonne of 
raw material/tonne cement. Therefore, the EF for quarry and mining becomes 0.0009 
tonne COi/tonne cement. 
Table 3.24 Production and Emissions From Quarry and Mining 



Production and Emissions 


Source: MoEF, FIMI, &Rauy and Reddy 


Production of material( million tonnes) 


703.1 


Total C02 (tonnes) 


1.46 


EF (tonnes C02/tonne material) 


0.002 



3.7.3 Coarse and Fine Aggregate Crushing. 

A report from the Central Pollution Control Board (2009) was used to calculated 
emissions for aggregate crushing. This report appears to be the only available report on 
aggregate crushing and it specifically focuses on stone crushers in Gujarat. The 
following labels are used to indentify coarse to fine aggregate sizes. 

• Coarse Aggregate - Black Trap (Metal or Kappchi) 

• Fine Aggregate - Black Trap (Grit or Dust) 



82 



Stone crusheres range from large to small. Large usually produce greater than 100 
tonne/hr (TPH) while small stone crushers produce 25 TPH. Stone crushing is higly 
labor intensive such that breaking, feeding, retrieving, and stockpiling are all performed 
manually. Table 3.25 shows the final fuel/energy emission factors calculated for stone 
crushing. 
Table 3.25: Emission Factors for Aggregate Crushing 



Fuel or Energy 


Emissions 


Electricity (tonne COi/tonne stone) 


0.002 


Diesel (tonne CCVtonne stone) 


0.0001 



3.7.4 Tranpsportation of Materials 

The assumption was made that all materials will be delivered by freight vehicles 

which are usually 3 axle trucks. For cement transportation the sources on emissions were 

gathered from Mckinsey and Company, 2010; Zhou, McNeil, 2009; and CMA, 2010. 

Fore aggregate transport the sources on emissions were granthered from Zhou, McNeil, 

2009; Reddy & Jagadish, 2003. Sources for fly ash transportation included Zhou, 

McNeil, 2009; Reddy & Jagadish, 2003. Tables 3.26 through 3.28 lists the emission 

factors for cement, aggregate and fly ash transportation and average distances travelled. 

Table 3.26 Emission Factors and Average Distance Travelled for Cement 
Transportation 



Cement 
Transportation 


Average 

kg CCVtonne- 

km 


AverageDistance 
Travelled (km) 


tonne 
CC>2/tonne 


Truck (Diesel) 


0.14 


280 


0.04 


Rail (Diesel) 0.008 


577 


0.005 


Sea (Diesel) 0.014 


900 0.01 



83 



Table 3.27 Emission Factors and Average Distance Travelled for Transport of 
Aggregate. 



Aggregate 
Transportation 


Average kg 

CCVtonne- 

km 


Average Distance 
Travelled (km) 


tonne CCVtonne 


Truck (Diesel) 0.14 







Table 3.28 Emission Factors and Average Distance Travelled for Transport of Fly 
Ash. 



Fly Ash (Truck) Average kg Average Distance tonne CC>2/tonne 
Transportation CO^tonne- Travelled (km) 
km 



Vanakbori 0.14 


302 


0.04 


Gandhinagar 0.14 


258 


0.035 


Sikka 0.14 


119.5 


0.016 


3.7.5 On-site Mixed Concrete 







The concrete mixer is usually located on-site where construction is occuring. The 
concrete mixer usually has a mechanical hopper attached as shown in Figure 3.8. 
Specifications regarding these types of concret mixers are shown in Table 3.29 which 
was gathered from various manufacturer specifications . 




Figure 3.8 Concrete Mixer with Mechanical Hopper 



84 



Table 3.29 Specifications of Concrete Mixer 


Specifications 


Units 




kW diesel engine 


7 


Average Productivity m /h 



Using the following equation the emission factor for can be calculated. The diesel 
emission factor was previously listed. Final emission factor was equal to about 0.00008 
tonne CCVtonne concrete. Information regarding specifications were gathered from 
various dealers. 



EF. 



EF„ 



1 



Energy 

^^x 

volume density of concrete 



■xEF 



diesel 



'kWh N 

V m 3 J 



1 



density of concrete 



tonnesCO, 



v tonne concrete /diesel 



3.7.6 Summary of Life Cycle Inventories 

A summary of emission factors are shown in Table 3.30. The largest contributing 
emission factor is arising from cement manufacturing. If the emission factors were 
compared to Reiners calculations for Ready Mixed concrete in Table 3.31, Indian on-site 
concrete mixing has lower emissions. Reiner reports a water emission factor, however, in 
Indian construction a bore well was used to gather mixing water but it was not clear if the 
well was manually dug or dug with diesel equipment. In this study the emissions for are 
much smaller than that shown by Reiner's study which is mainly because aggregate 
crushing in India more manually labor intensive. 



85 



Table 3.30 Summary of Emission Factors Leading Up to Concrete Mixing. 



Name 


Material 


Total Energy 

(MJ/tonne 

material) 


EF(tonne 

C0 2 /tonne 

material) 


Quarry and Mining 


sand, clay, etc. 


4.7 


0.002 


Cement 


cement 


3340 


0.84 


Aggregate Crushing 


coarse & fine 
aggregate 


12 


0.0024 


Transportation 


cement (truck) 


1.85 


0.038 


cement (rail) 


0.12 


0.005 


cement (waterway) 


0.19 


0.01 


fine agg. (truck) 
coarse agg (truck) 


1.85 
1.85 


0.01 
0.01 


fly ash (truck) 


1.85 


0.03 


On-site mixing 


concrete 




0.00008 



Table 3.31 Reiner's (2007) Emission Factor Calculations for R eady Mixed Concrete 

U.S. 



Material 



Sector Name 



Total Total tonne 

Energy, C0 2 e/tonne 

MJ/tonne of material 



. . Sand, gravel clay, „ 

Virgin Aggregates B y 110 0.008 

refractory mining 



Cement 



Cement 
manufacturing 



8030 



Transportation 



Truck 
Transporation 



2.54 



Rail Transporation 0.41 



1.21 



0.0003 



0.00003 



Ready Mix 



Ready-Mix 

Concrete 

Manufacturing 



205 



0.019 



Mix Water 



Water, sewage, and 
other systems 



0.003 



86 



3.8 MFA-LCA of Cement Use in Rajkot 

A regional material flow analysis is meant for tracking the flow of materials that arrive at 
a specific region and are used in that region. The equation below shows the calculation 
for an MFA-LCA. 

MFA- LCA Emissions = Total Material x EF^ 

Where, 

EF L ci= emission factor from a certain material 
A report by Chavez, Ramaswami, Dwarakanath, Guru & Kumar. (2012) was one of the 
first MFA-LCA for a city in a developing country that determined an MFA-LCA for 
cement use in Delhi. The cement emission factor (EF) did differ from that calculated for 
Rajkot's cement. The EF was 0.93 tonnes CC^/tonne cement but was based on 1994 
data from Hendricks et al (2004). The material flow analysis however was gathered 
from the Cement Manufacturers' Association (CMA). The CMA reports cement use per 
year for all states and the city of Delhi. However, cement use in Rajkot was gether from 
a personal communication with an Ambuja representative (201 1). An estimated 45,000 
tonnes per month are designated for the city of Rajkot. The cement is dispursed for trade 
(small businesses) and non-trade (large construction) use. Therefore approximately 
540,000 tonnes of cement are used per year. This information is shown in Table 3.32. 
Using the EF calculated for cement, emissions from cement use in Rajkot per year is 
about 453,600 tonnes/yr. 



87 



Table 3.32 Information Regarding Rajkot Cement Use and Total Emissions per 
Year 



Trade 



20,000 tonnes/month 
540000 tonnes cement /yr 
MFA-LCA = 453600 tonnes C0 2 /yr 



Non-trade 



25,000 tonnes/month 



In Chapter 2 the total population was reported to be about 1.4 million. Therefore, 
percapita cement use is about 0.39 tonnes of cement/person. Chavez et al. (2012) 
determined Delhi's percapita cement use to be about 0,24 while in Denver Reiner (2007) 
calculated about 0.50. 

3.9 MFA-LCA for Concrete Mixtures in Rajkot 

Since exact material flows of aggregate are unknown for Rajkot it might be better to 
determine how much emissions a cubic meter of certain concrete mixtures used in Rajkot 
would result from using these mixtures. The calculation of an MFA-LCa for a concrete 
mixture can be determined as follows: 



MFA- LCA Emissions = ^ 



( total material 
V volume 



:EF, 



LCI 



An MFA-LCA was determined for a conventional M35 concrete pavement mixture and 
compared to pervious and HVFA concrete mixture. Table 3.33 shows the MFA data 
while Table 3.34 shows the LCA data. Table 3.35 shows the key material cement and 
how it changes for each mixture as well as final MFA-LCA values. Pervious concrete 
provides a 21% reduction in emissions per cubic meter, while HVFA concrete provides 
about 47% reduction in emissions per cubic meter when compared to a traditional 
concrete pavement mixture. 



88 



Table 3.33 MFA Data for M35, Pervious and HVFA Concrete Mixtures 



MFA 




M35 


Pervious 


50% Fly Ash 


M ate rial/Prope rty 




kg/m concrete 




Cement 


400 


311.5 


196 


Fly Ash 








196 


Coarse Aggregate 


1271 


1704 


1022 


Fine Aggregate 


704 


110 


775 


Estimated Density of 
Concrete 


2535.0 


2236.3 


2363.2 


w/c 


0.4 


0.3 


0.4 



MFA xLCA = EMISSIONS 



Table 3.34 LCA Data and Total Emissions Calculations from an MFA-LCA on Concrete Mixtures 



LCA 


Total Emissions 


Name 


Material 


Emission Factors 


M35 


Pervious 


50% Fly Ash 


tonne C0 2 /tonne material 


(tonne C0 2 /m ) 


Quarry and Mining 


cement 


0.0009 


0.00037 


0.00029 


0.00018 


Cement 


cement 


0.84 


0.33600 


0.26166 


0.16464 


Aggregate Crushing 


coarse & fine aggregate 


0.0024 


0.00481 


0.00442 


0.00438 


Transportation 


cement (road) 


0.0384 


0.01535 


0.01196 


0.00752 


line aggregate (road) 


0.0103 


0.00724 


0.00113 


0.00797 


coarse aggregate (road) 


0.0103 


0.01307 


0.01752 


0.01051 


fly ash (road) 


0.0310 


0.00609 


On-site mixing 


concrete 


0.00008 


0.00020 


0.00018 


0.00019 


Total (tonne C02/m 3 ) 


0.38 


0.30 


0.20 



00 



Table 3.35 Cement Material Content and MFA-LCA Emissions for Certain 
Concrete Mixtures 


Key Material and Overall Emissions 


Traditional 


Pervious 


HVFA (50%) 


Cement (kg/m3) 400 312 196 


Concrete Mixture MFA-LCA 
(tonne C02/m 3 ) 


0.38 


0.30 


0.20 



3.10 Summary 

The CMA reports a cement emission factor (0.83 tonnes COi/tonne cement) very close to 
the calculations performed in this disseration for the state of Gujarat (0.84 tonnes 
CCVtonne cement). However, it was necessary to perform the cement life cycle 
inventory because there are other contradicting sources reporting a range of emission 
factors for Indian cement (0.6 to 1.0 tonnes CCVtonne cement). Actually this range is 
represenative of the how large companies and small companies have generate a majority 
of the electricity on-site. However, many of the efficiency of production for the smaller 
companies are less than that of the larger companies which seems to make the emission 
factors fluctuate. Other materials and transportation needed for concrete revealed that 
much of the emissions is arising from cement manufacturing. An MFA-LCA of cement 
in Rajkot revealed per capita cement use in Rajkot is 0.15 tonnes/person more than Delhi 
but is still 0.1 1 tonnes/person below that of a U.S. city like Denver. Final MFA-LCA 
calculations for pervious concrete and HVFA concrete mixtures showed at most a 21% 
and 47% reduction in emissions, respectively, compared to a conventional concrete used 
in Rajkot. 



90 



4. Stormwater Solution Demonstration with Pervious Concrete: Structural and 
Environmental Tests 

In order to complete a pervious concrete system demonstration in Rajkot, India it 
was necessary to decide on an acceptable mixture and system design. Chapter 4 
introduces the initial laboratory testing that was performed before commencing the study 
in Rajkot, India. Phase I of laboratory testing was based on mixture proportioning 
established through Hager's study (2009). Upon successfully achieving a design strength 
of 13.8 MPa (2000 psi) through Phase I testing, the mixture design was applied in Rajkot, 
India. The testing and results gathered from the pervious concrete project implemented in 
Rajkot is discussed in Chapter 4. Throughout the discussion on Rajkot it is explained that 
plans changed while on-site and the pervious concrete field implementation changed into 
a small demonstration project. A perspective on international collaboration is also given 
in Chapter 4. The results gathered in Rajkot led to Phase II laboratory testing. The 
reality of having compulsory changes during the project such as curing techniques, and 
having different aggregate shape, and the shape of the test specimen are just a few 
reasons why a second phase of lab testing was needed in this study. The second phase of 
the study demonstrated that the effectiveness of a concrete technology (such as pervious 
concrete) serving as a climate adaptation solution is dependent on mechanical measured 
properties (i.e. compressive strength). The results of mechanical properties can also 
affect the decision of whether to use the technology in cities despite the technology 
satisfying other expected benefits (i.e. porosity, filtration). 



91 



4.1 Study Design and Laboratory Phase I Testing 

As stated in Chapter 1 and 2 Rajkot Municipal Corporation and ICLEI South Asia 
were interested in adopting material interventions that would lead to helping the city 
develop urban infrastructure design for carbon mitigation and climate adaptation as well 
as helping the city meet some of its lacking infrastructure needs. It was identified from 
Rajkot' s development plan (as discussed in Chapter 2) that the city needed improvements 
in stormwater management. There was either very little to no stormwater infrastructure 
existing in the city. Initially, a field demonstration of a pervious concrete system in 
Rajkot, India was agreed upon. A visit to Rajkot, India was made in January 201 1 such 
that representatives from the University of Colorado Denver (Dr. Stephan Durham), 
ICLEI-South Asia (Ms. Laasya Bhagavatula), Rajkot Municipal Corporation (Ms. Alpana 
Mitra and Mr. Mitesh Joshi) and Lakhlani Associates (Mr. Jayant Lakhlani) met and 
discussed the various projects to be researched in Rajkot. During this initial visit the site 
for the pervious concrete system was chosen. Figure 4.1 shows the initial site chosen for 
the pervious concrete system test section in Rajkot. The site was located at the Raiya 
wastewater treatment facility where they were placing a high volume fly ash (HVFA) 
concrete road. The size of the site was to be 3.5 m x 15 m (1 1.5 ft x 49 ft). The slope, 
location of drainage, and the availability of materials and equipment on site were reasons 
for choosing this area for the pervious concrete placement. 



92 




Figure 4.1 Proposed Pervious Concrete System Site 

Chapter 1 of this dissertation mentioned that the existing literature on pervious 
concrete did not discuss the transfer of pervious concrete mixture and system designs, 
that had been researched in the United States, to countries where not only materials 
(aggregate, water, and cement) differed, but construction techniques as well. Rajkot 
presented an opportunity to test a pervious concrete system. The same system and 
mixture proportioning that had been successfully used in a field demonstration in Denver, 
Colorado, as a parking lot pavement on the Auraria Campus, (Hager, 2009) would be the 
model for the pervious concrete system pavement section to be placed in Rajkot. Figure 
4.2 shows the parking lot pervious concrete system pavement placed on the Auraria 
Campus. In Hager's study, six asphalt parking stalls were replaced with the pervious 
concrete pavement system. The pervious concrete pavement is blatantly called a system 
because there are two other important layers below the pervious concrete pavement. The 
layers usually are coarse aggregate directly below the pervious concrete and then fine 
aggregate. If the drainage design consists of a perforated pipe, then another layer of 
coarse aggregate is used to fill the trench where the perforated pipe sits (Refer to Figure 

93 



4.3). The study by Hager (2009) used alternative materials (i.e. waste products) to 
replace the coarse aggregate and fine aggregate. As seen in Figure 4.3 coarse aggregate 
was replaced with recycled concrete and the fine aggregate was a mixture of sand and 
crushed glass. Above and below the crushed glass/sand was a geotextile fiber that 
prevented the sand from entering into the other layers. The system design for Rajkot 
would not consist of the waste products as used in Hager' s study, but the alternative 
design presented by Hager is a good model demonstrating the recycling of waste 
products. Additionally, Hager's results from laboratory testing demonstrated that a fine 
aggregate content with at most 7.5% of total weight of aggregate, a cementitious 
materials content between 311 kg/m 3 (525 lb/yd 3 ) and 326 kg/m 3 (5501b/yd 3 ) , maximum 
20% Class F fly ash, and a water/cementitious ratio of 0.30, improved the pervious 
concrete's resistance to freezing and thawing cycles, maintained approximately 10% 
porous structure, and helped meet a design strength of 13.8 MPa (2000 psi) (Hager, 
2009). 




Figure 4.2 Pervious Parking Lot Pavement on Auraria Campus in Denver, Colorado 



94 



Woven geotextile as specified on this drawing 




Install 16 MIL (0.016 in, 0.41 mm; 

impermeable membrane under 

pipe & wrap it to within 1 in 

(2.54 cm) of top of pavement 

surface to serve as horizontal 

flow barrier. 



4 in (10.16 cm) diameter 
schedule 40 HDPE 
perforated underdrain pipe 

Fill trench with same 
recycled concrete as 
used 



Monolithically placed pervious concrete pavement 
(Mix of AASHTO #8 gravel, portland cement, fly ash 
and admixtures per specifications in Chapter 5) 



"(15.24 cm) 

D = 8" (20.32 cm) 
{ 1" (2.54 cm) 
6" (15.24 cm) 



Woven geotextile fabric Meeting: 

ASTM D4751-AOS US Std. Sieve #50 to #70, 

ASTM D4533 min. Trapezoidaltear strength 1 00 x 50 lbs, 

Minimum COE Specified open area of 4%. 

* Base Course: AASHTO #67 sized Recycled concrete 
** Sand I Crushed Glass mixture: 

90% sand, 10% crushed glass both ASTM C 33 gradation 



Figure 4.3. Details of the Pervious Concrete System for the Parking Lot Installation 
(Hager, 2009) 

4.1.1 Material Properties 

A Type I/II ordinary portland cement from Holcim, Inc. was used for Phase I 
laboratory testing. The specific gravity of the cement was 3.15. Chemical and physical 
properties of the cement are shown in Table 4.1 and 4.2. 

The maximum size of coarse aggregate used in the mixture design in the study by 
Hager was 9.5 mm (3/8 in). However, the availability of aggregate size was unknown for 
Rajkot. As such the size of aggregate used in this study was based on well-graded 
aggregate that ranged from a maximum aggregate size of 25.4 mm (1.0) in to a nominal 
maximum aggregate size of 19 mm (0.75 in). Aggregates were provided by Bestway 
Aggregate in Colorado. The coarse and fine aggregate both met American Society for 
Testing and Materials (ASTM) C33. The coarse aggregate met ASTM size number 57 
and 67 gradation. A sieve analysis was performed by WesTest in Denver, CO for both 
the coarse and fine aggregate. The results of the analysis are presented in the Appendix 
as Figure C.l and Figure C.2. The specific gravity for the coarse aggregate was 2.61 with 



95 



an absorption capacity of 0.6%. The specific gravity for the fine aggregate was 2.63 with 

an absorption capacity of 0.7%. 

Table 4.1 Chemical Properties of Cement along with Standard Limits 



Chemical Property 


Holcim Results (%) 


ASTM C 150 Limits 


Si02 


19.7 




A1203 


4.6 


6.0 max 


Fe203 


3.2 


6.0 max 


CaO 


63.8 




MgO 


1.3 


6.0 max 


S03 


3.2 


3.0 max 


Loss on Ignition 


2 


3.0 max 


Insoluble Residue 


0.53 


0.75 max 


C02 


1.1 




Limestone 


2.9 


5.0 max 


CaC03 in Limestone 


83 


70 min 


Inorganic Processing Addition 





5.0 max 


C3S 


62 




C2S 


9 




C3A 


7 


8 max 


C4AF 


10 




C3S+4.75C3A 


95 


100 max 



Table 4.2 Physical Properties of Cement Along with Standard Limits 



Physical Properties 


Holcim Results 


ASTM C 150 Limits 


Air Content (%) 


7 


12 max 


Blaine Fineness (m2/kg) 


398 


260 < x < 430 


Autoclave Expansion (%) (C151) 


0.01 


0.8max(C151) 


Compressive Strength MPa (psi) 




C150 


3 days 


31.4(4450) 


10.0 (1450) min 


7 days 


38.7 (5620) 


17.0 (2470) min 


Initial Vicat (minutes) 


134 


45-375 


Mortar Bar Expansion (%) (C 1038) 


0.003 




Heat of Hydration: 7 days, kJ/kg (cal/g) 


343 (82) 





Hager's study showed that 20% Class F fly ash could be used in pervious concrete. 
Although the fly ash decreased the compressive strength of the pervious concrete 
mixtures, it was assumed that long-term strength would either be greater than or equal to 



96 



that of a 100% ordinary portland cement pervious concrete mixture (Hager, 2009). 
Although Hager's mixture design approach was going to be tested in Rajkot the decision 
was made not to use fly ash in any of the mixtures designs. In fact, certain engineering 
officials in Rajkot were emailed, before the project was performed in Rajkot, regarding 
fly ash properties, but no information was provided. Thus, due to the possibility of 
decreasing compressive strength with the use of fly ash and the unknown properties of 
the sources of fly ash in Rajkot, no fly ash was used throughout the pervious concrete 
study. 

An air entraining admixture was used in Hager's study to aid in freeze-thaw 
resistance. However, in Rajkot freezing is not a concern. The temperature, as discussed 
in Chapter 2, in Rajkot is usually hot. In addition, a hydration stabilizing admixture was 
used in Hager's study in order to prevent an impervious zone. The impervious zone 
usually forms when all the paste and aggregate settle to the bottom of the placement of 
the pervious concrete. However, it was very likely that this type of admixture was not 
readily available in the city of Rajkot. No admixtures were used in the pervious concrete 
study. 
4.1.2 Mixture Design 

Table 4.3 lists the mixture designs for Phase I. Although 10% or more porosity 
was estimated by Hager from the voids present in the concrete mixture, there was no 
correction made to the assumed air content for the mixture design. Thus, Mixture 1 of 
Phase I testing assumed an air content of 2% as Hager did in 2009. However, a decision 
was made to assume an air content of 13% after testing for percent porosity of Mixture 1 
(percent porosity is discussed in the results of the Phase I testing). Mixture 2 was 



97 



assumed to have 13% air content. Both mixture designs consisted of a cement content of 
311 kg/m (525 lb/cy ), water-cement (w/c) ratio of 0.3, and a fine aggregate content of 
6% of total weight of aggregate. 
Table 4.3 Mixture Proportions for Phase I Laboratory Testing 











Coarse 


Fine 








Water, 


Cement, 


Aggregate, 


Aggregate, 


Air 


Mixture 


W/C 


kg/m 


kg/m 


kg/m 


kg/m 


Content, % 


1 


0.3 


93 


311 


1946 


110 


2 


2 


0.3 


93 


311 


1662 


106 


13 


1 kg/m' = 


1.6856 lb/y 


d J 











4.1.3 Test Methods 

During Hager's study the ASTM Subcommittee C09.49 was in the process of 
developing standards for testing permeability, compressive strength, flexural strength, 
fresh and hardened concrete density, void content, and porosity. The ASTM 
Subcommittee C09.49 developed a standardized test for determining the density and void 
content of freshly mixed pervious concrete in 2008 referred to as ASTM C1688. ASTM 
C1688 was not used in this study; a couple of reasons being that ASTM C1688 was fairly 
new and could be revised and it was the desire of the author to try and mimic the 
preferred field compaction method of using a roller. ASTM CI 688 requires the use of a 
standard proctor hammer to compact the pervious concrete. Standardized methods for 
compressive strength are still a work in progress by ASTM Subcommittee C09.49. As 
such, ACI committee 522 still refers to ASTM C39 for compressive strength testing of 
pervious concrete samples, but ACI makes note that a better compressive strength test is 
needed. Standard procedures for preparing and curing pervious concrete samples and 
tests for porosity and hydraulic conductivity have yet to be established as well. The 



98 



procedures for preparing and testing samples in this study are described in the next few 
paragraphs. 

Laboratory batching of pervious concrete is not discussed in ASTM nor ACI 
documents. Based on the procedures described by Hager (2009) and Tennis, Leming, & 
Akers (2004) the batching mostly followed the same procedures used for conventional 
concrete. However, mixing time was mostly dependent on the consistency of the 
mixture. The consistency of the mixture is best described in Tennis, Leming, & Akers 
(2004). The method of checking consistency helps determine if the water content is 
controlled. A handful of mixed pervious concrete is taken into the hand and shaped into 
a ball. If the mixture partially remains in the shape of a ball, but leaves a lot of paste 
residue on the hands or void structure is hindered by too much paste, then the mixture is 
"too wet". The mixture consistency should be somewhere between "too dry" and "too 
wet" such that the mixture remains in the shape of a ball and this is called "proper 
amount of water". Figures 4.4 (a) through 4.4 (c) is used to depict the three mixture 
consistencies. 




(a) (b) (c) 

Figure 4.4 Mixture Consistency (a) Too Dry, (b) Proper Amount of Water, (c) Too 
Wet (Tennis, Leming, & Akers, 2004) 

Immediately after mixing was finished the pervious concrete was placed in the molds and 

the following procedure was used making specimens and curing specimens. 



99 



Method of Preparing Specimens and Curing 

1. Concrete was placed into 10.2 cm x 20.3 cm (4 in x 8 in) cylinders using 2 lifts. 

2. Each lift was rodded 25 times 

3. After rodding each layer the outsides of the mold were tapped with a mallet (or 
hand for plastic cylinder molds) 10-15 times. 

4. The last lift was added such that approximately 3.2 mm (1/8 in) to 12. 7mm (0.5 
in) of concrete was above the rim of the mold just before compaction. The layer 
was then compacted with a rolling pin that weighed about 2.8 kg/m (1.9 lb/ft). 
While applying my weight over the rolling pin, the actual weight being applied 
over the surface of the specimen was about 29.8-37.2 kg/m (20-25 lb/ft). The 
rolling pin was rolled over the surface until no more settlement was apparent. 

5. The specimens were immediately covered with 6 mil (0.006 in, 0.15 mm) plastic. 
The plastic was sealed with tape. 

6. The specimens were placed into a curing room that remained at a fixed 
temperature of23±2"C(73±3°F) and humidity at about 55%. 

7. The specimens were cured for 14 days before they were removed from the molds. 
While curing the concrete specimens were sprayed everyday for 14 days as part 
of the curing process. Spraying with water was done to mimic curing pervious 
concrete out in the field. 

Note: 15 cm (6 in) cubes and 25.4 cm x 25.4 cm x 17.8 cm (10 in x 10 in x 7 in) block 
samples were made in addition to the cylinders. The cube was placed using 3 lifts, 
rodding each lift 18 times, and the outside of the cube 10-15 times with a mallet after 
each lift was consolidated. The last lift was compacted with the rolling pin. Cubes were 
an essential testing element because in India cubes are the preferred shape to be tested 
for compressive strength. The procedure for the making the block was similar to the 
cylinders and cubes except 4 lifts were used and each lift was rodded 50 times. Cores 
having 7.6 cm diameter x 17.8 cm lengths (3 in x 7 in) were drilled from the blocks. 
Hager hypothesized that compressive strength from cores best represented field 
strengths. 



100 



The reason the rolling pin was used for compaction was because in the field a 
roller usually weighing between 44.6 to 59.5 kg/m (30 to 40 lb/ft) or 0.44 kg/cm (2.5 
lb/in) is used in the compaction. The rolling pin with weight applied was considered 
satisfactory for these samples especially if compressive strength was met. 

The method of determining compressive strength involved ASTM C39 with some 
modifications. The loading rate used for the cylinders was about 0.08 + 0.05 MPa/s (12 + 
7 psi/s) and for the cubes the rate was approximately 0.04 + 0.05 MPa/s (6 + 7 psi/s). 
Lower loading rates, as compared to ASTM C39, were used because it was unclear 
whether the voids in the pervious concrete affect the ultimate strength at different loading 
rates; therefore it was assumed a lower loading rate would provide a necessary caution. 
Additionally, ends of the specimens were sawed off if the testing surface of the samples 
needed to be level. At the most 1.3 cm (1/2 in) was sawed. The samples were then tested 
between two neoprene pads. The cubes, on the other hand, were tested between to steel 
plates. Figure 4.5 shows the difference between testing techniques for the cylinders and 
cubes. 




(a) (b) 

Figure 4.5 Compressive Strength Testing (a) Using Neoprene Pads for Cylinders 
and (b) Steel Plates for Cubes 

Determining the percentage porosity of the pervious concrete is important towards 

the estimation of storage capacity (portion of the concrete that can be filled with rain). 

101 



Tennis, Lemings, and Akers (2004) gave the example if the concrete has 15% porosity 
then a 25.4 mm (1 in) thick pervious concrete pavement could store 3.8 mm (0.15 in) of 
rain before the rain would no longer be stored and become runoff. Typical porosity can 
range between 15% and 25% (Tennis, Lemings, and Akers, 2004). The following 
method was used to determine percent porosity: 
Method of determining percent porosity 

1. Dry concrete sample 

2. Weigh the dry concrete sample and record the value ( W S( ± vy ) 

3. Weigh an empty container and record the value (W c ) 

4. Fill the empty container with water to a certain level and call this the initial level 
(i.e. 20 cm from the bottom of the container) 

5. Weigh the filled container and record the value (W c+W ) 

6. Determine the mass of water in the container (W c+W -W c = W W [) 

7. Place the dry sample in the filled container (approx. 5 min) 

8. Empty the water from the filled container until the water level is at the initial level 

9. Weigh the filled container with sample and record the value (W c+W+S ) 

10. Determine the mass of the water in the container with the sample (W c+w+ii -W c - 
W sdry = W w2 ) 

11. Determine the mass of the water displaced by the sample's solids (W w i-W W 2 = 
W w3 ) 

12. Convert mass of the water displaced to volume of the water displaced by dividing 
by the density of water (V w ) 

13. Determine the volume of a solid sample based on sample dimensions (V ss ) 

14. Determine the percentage voids [(V ss -V w )/V ss x 100= P Y ] 

15. Cross check the calculation of percentage voids by determining the mass of water 
emptied from the container in step 8. (W W 4) This also represents the mass of water 
displaced by the solids from the sample 



102 



16. Convert the mass of water from step 15 into volume of water displaced by 
dividing by the density of water (V^a) 

17. Determine the percentage voids [(V^-V^/V^ x 100= P s i]. 



Hydraulic conductivity (also known as permeability and infiltration rate) is the flow 
rate through the concrete. According to Tennis, Leming, and Akers (2004) typical flow 
rates are 0.2 cm/s (288 in/hr) or higher. Hydraulic conductivity is important when 
designing the pervious concrete system for stormwater management. However, the 
permeability of the pervious concrete is not the controlling factor; it is also necessary to 
know the permeability of the subgrade soils that the pervious concrete system will be 
placed on (Tennis, Leming, and Akers, 2004). In this study, however, only the method of 
determining hydraulic conductivity of the pervious concrete system was discussed. The 
procedure used for determining hydraulic conductivity is somewhat based on the falling 
head method. The test was adapted from Delatte, Miller, and and Mrkajic (2007). The 
test is described as follows: 

Method of determining Hydraulic Conductivity 

1. Use a 10.2 cm x 20.3 cm (4 in x 8 in) cylinder mold that has a 1.9 cm (3/4 in) hole 
drilled through the bottom of the cylinder. Either foam rubber or plumber's putty 
is secured to the bottom of the cylinder so as not to allow water to flow away from 
the cylinder when the cylinder is filled with water (See Figure 4.6) 

2. Saturate the sample (i. e. make sure samples have been moistened completely) 

3. Plug the hole of the cylinder with a stopper (make sure a rod or chain is attached 
to the stopper in order to pull it out without much disturbance to the water in the 
cylinder). 

4. Place the testing apparatus (cylinder) with plumber's putty adhered to the 
cylinder and surface of the sample. 

5. Fill the cylinder with water until a near spherical shape of water forms at the top 
of the cylinder. 



103 



6. Pull the stopper out of the cylinder while initiating a stop watch which is used to 
record the time it takes for the water to drain from the cylinder. 

7. Stop the stop watch once no water is seen draining from the 1.9 cm (3/4 in) hole. 

8. To calculate the water the following equation is used 

ah (h^ 



aL //ii \ 
At \h 7 ) 



Where, 

k = hydraulic conductivity (length/time) 

a = cross-sectional area of cylinder (not 1.9 cm (3/4 in) hole)(length ) 

A = cross-sectional area of sample (length ) 

L = length of sample (length) 

t = total time to for water to drain from cylinder (time) 

hj= initial water level (length) 

hi= final water level (length) 

In = natural logarithm 

9. Although the cylinder is allowed to drain completely, /?2 is not exactly zero. Some 
level of water is left near the bottom of the cylinder and can be measured by 
pouring into a graduated cylinder. The volume of water in the graduated cylinder 
is divided by the area of the cylinder to get an approximate height of the wate, r 
which is h2. 




(a) (b) 

Figure 4.6 Hydraulic Testing Apparatus (a) Cylinder with Stopper and Putty (b) 
Hole Drilled in Cylinder for Draining Water from the Cylinder into the Pervious 
Concrete 



104 



4.1.4 Phase I laboratory results 

Density 

The fresh concrete density was determined very similarly to the procedure 
described in ASTM C138, however, the last layer of pervious concrete was compacted 
using the rolling pin. Density for Mixture 1 was accidently not recorded. The density for 
Mixture 2 was recorded as 1922 kg/m (120 lb/ft). According to Tennis, Leming, and 
Akers (2004) typical unit weights range between 1600 kg/m 3 and 200 kg/m 3 (100 lb/ft 3 
and 125 lb/ft ). Mixture 2 falls within these typical unit weights. Unit weights are 
considered to be a proof of whether mixture proportions are consistent (i.e. quality 
control) especially when it comes to jobsite mixture deliveries. 
Porosity 

Both mixtures were tested at 7 days of curing. The samples were supposed to be 
tested at 28 days of curing, however, the project in Rajkot had to be scheduled during the 
curing days of Mixture 1 and Mixture 2 so Mixture 1 was cured up to 23 days and 
Mixture 2 was cured up to 15 days. Table 4.4 shows the percent porosity recorded for the 
samples. 
Table 4.4 Porosity of Samples from Mixture 1 and Mixture 2 (Reported in Percent) 



Curing Days 


Samples 


Cylinders 


Mixture 1 
Cores 


Cubes 


Cylinders 


Mixture 2 
Cores 


Cubes 


7 




1 

2 


13 
13 


35 
26 


23 


20 
20 


33 
39 


23 
23 


23*/15** 




1 

2 


17 
17 


26 

34 


14 
19 


18 
11 


28 

32 


26 

27 


Average 


15 


30 


19 


17 


33 


25 



* Mixture 1 cured up to 23 days 
** Mixture 2 cured up to 15 days 



105 



Table 4.4 does show differences in percent porosity between curing days. 
However, the percent porosity was not expected to change by much between curing days. 
By at least 7 days the cement paste should have hardened and movement perhaps should 
have been limited to micro movements. Changes in % porosity between curing days 
would most likely be due to random orientation of aggregate especially if non-uniform 
graded aggregate is used. In addition, Table 4.4 shows how the cored cylinders for both 
Mixture 1 and Mixture 2 resulted in a higher percent porosity compared to the other 
cylinders and cubes. This could be an example of how a sample with greater area may 
not get rodded as well as a smaller sample. Although there were some differences in % 
porosity among the different samples the recorded values fall within the typical range of 
15% to 25% according to Tennis, Leming, and Akers (2004). A side by side comparison 
of the samples is shown in Figure 4.7. From Figure 4.7 the voids within the samples are 
visible. 




Figure 4.7 A Side by Side Comparison of the Pervious Concrete Samples 

Hydraulic Conductivity 

In Phase I lab testing the hydraulic conductivity was determined only using the 
large pervious concrete blocks before coring them (as seen in Figure 4.6). Using the 
Equation described in the "Method for determining hydraulic conductivity" the values 



106 



recorded for Mixture 1 and 2 are listed in Table 4.5. A value was accidently not recorded 
for Mixture 1 at 7 days of curing. Table 4.5 represents the average of at least three tests 
performed on the samples. The test results, overall, yielded a range of hydraulic 
conductivities from 0.22 cm/s (0.09 in/s) to 0.41 cm/s (0.2 in/s), which is higher than the 
lowest typical flow rate of 0.2 cm/s (0.08 in/s) as reported by Tennis, Leming, and Akers 
(2004). Originally, Delatte, Mrkajic, and Miller (2009) called this method of determining 
hydraulic conductivity the drain time test. Delatte, Mrkajic, and Miller (2009) performed 
drain time tests in several locations where pervious concrete was placed as parking lots 
and sidewalks. Cores were taken from these locations and falling head tests were 
performed on these cores. Delatte, Mrkajic, and Miller's work resulted in a good 
correlation between drain time test and hydraulic conductivity. Because the various sites 
they visited were based on different mixture designs and most likely aggregate gradation 
the authors suggested that the correlation between drain time test and falling head test be 
rationalized through the following empirical formula. 

k = 2533 x e -° 062t 

Where, 

k = hydraulic conductivity from laboratory tests (in/hr) 

t = drain time test (s) 
If the average drain time values ranged from 27 s to 44 s for Mixture 1 and 2 then by 
Delatte, Mrkajic, and Miller's empirical formula the hydraulic conductivity ranges from 
0.34 cm/s (475 in/hr) to 0.12 cm/s (166 in/hr). This range is close to the values reported 
in Table 4.5, therefore, either relating the test method values directly to the falling head 
test or the empirical formula is appropriate. 



107 



Table 4.5 Average Hydraulic Co nductivity for Mixture 1 and 2 

Curing Mixture 1 Mixture 2 
Days (cm/s) (cm/s) 

7 - 022 

23*/15** 0.41 0.36 

* Mixture 1 cured up to 23 days 
** Mixture 2 cured up to 15 days 
1 cm/s = 0.3937 in/s 

Compressive Strength 

All recorded values for compressive strength are shown in Table 4.6a and Table 
4.6b and the average compressive strengths are displayed in the Figure 4.8. The design 
strength was 13.8 MPa (2000 psi). By 7 days of curing both Mixture 1 and 2 
demonstrated that cylinders and cubes were nearing design strength. For this study it was 
desired to have the 10.1 cm x 20.3 cm (4 in x 8 in) cylinder as the reference for strength, 
meaning that the strengths of cubes and cores could be related with a strength ratio or 
factor. According to Mindess, Young, and Darwin (2003) the common cube to cylinder 
strength ratio is 1.25. This factor means that the cube strength is usually higher than the 
cylinders. However, upon comparing cylinders and cube compressive strengths for both 
Mixture 1 and Mixture 2, the cube compressive strengths were lower than the cylinder 
strength. At this point in the study, a strength conversion factor used to attain an 
equivalent cylinder compressive strength for cubes was not determined. 

For the 7.6 cm x 17.8 cm (3 in x 7 in) cores the length to diameter ratio is 2.3. 
ASTM C39 only lists strength conversion factors for length to diameter ratios equal to or 
less than 1.75. But, Mindess, Young, and Darwin (2003) did provide a compressive 
strength factor equal to about 1.03 which represented the cylinder to core ratio. This 
meant the resulting core compressive strengths are usually lower than the cylinders. 



108 



By the 23 r and 15 £ day of curing, the cores had increased in strength and if 
multiplied by the 1.03 factor the average values would be about an equivalent cylinder 
value equal to 12.3 MPa (1784 psi) and 13.3 MPa (1929 psi) for Mixture 1 and Mixture 
2, respectively. However, cubes compressive strengths did not increase. Figure 4.8 
shows two trends for the cubes. Compressive strength results, for Mixture 1 cubes, 
decreased by as much as 7 MPa (1015 psi). While Mixture 2 cubes had mostly remained 
consistent with 7-day compressive strengths, both mixtures never reached 13.8 MPa 
(2000 psi) according to cube compressive strength results. The cylinders, however, did 
reach and pass the design strength by 15 days of curing. The relationship between the 
cylinders and cubes was still in question but the success of the cylinders passing design 
strength by as much as 2.2 MPa (319 psi) was satisfactory for producing pervious 
concrete in Rajkot, India. Besides, a trial mixture would be completed in Rajkot, India 
before proceeding with the field placement. 

Table 4.6a Mixutre 1 Compressive Strength Results 



Mixture 1 (MPa) 
Sample cylinders 
7-days 23-days 



1 

2 
Average 
Std. Dev. 



11 
16 

14 
3 



14 
18 
16 

3 



Mixture 1 (MPa) 

cores 
7-days 23-days 




Mixture 1 (MPa) 

cubes 
7-days 23-days 



13 
12 
12 




6 
11 

8 

4 



lMPa= 145.038 psi 



109 



Table 4.6b Mixture 2 Compressive Strength Results 





Mixture 2 (MPa) 


Mixture 2 (MPa) 


Mixture 2 (MPa) 


Sample 


cylinders 




cores 




cubes 




7 


15 


7 




15 


7 15 


1 


13 


17 


8 




13 


9 6 


2 


13 


16 


9 




12 


6 8 


Average 


13 


16 


9 




13 


7 7 


Std. Dev. 








1 




1 


2 2 



lMPa= 145.038 psi 



Compressive Strength (MPa) 






,o 


1 MPa =145.038 psi 


y^ — — ^ 






-9- M 1 _cy linders 




-H-M^cores 
^*^ -*-Ml cubes 


h / / 


-o- M2_cylinders 


a / ' 

H / ' 


-a- M2_cores 
-A- M2_cubes 


B // 
III 
III 


III 


) 7 14 21 28 
Days 



Figure 4.8 Average Compressive Strengths for Mixture 1 and Mixture 2 

Some observations made during the compressive strengths tests revealed that the 
cylinders and cubes fractured in certain patterns. For example, Figures 4.9a and 4.9 b 
show at least two paths the fracturing took during compressive strength testing of the 
cylinders. The fracture paths that most commonly occurred for the cubes during the test 
are shown in Figure 4.10. In the figures, a generalized vertical line is used to represent 
some of the fracture paths but the actual paths followed the location of the voids and 
paste surrounding the aggregates. There were examples where the fracture occurred 



110 



through the aggregate as seen in Figure 4.11. If the fracture occurred through the 
aggregate this is could be a good indication of the bond between the paste and aggregate. 



Ok 

m 


IT- 


_ 


t«0 • i 

— -1 ■ 










(a) (b) 

Figure 4.9 Fracture Paths for Cylinder Pervious Concrete Samples 







< * - 2 




' /■ ■ 


tfc§l 






^■■F 


i>*i 




<' l "* 




' ,'j 



Figure 4.10 Fracture Paths for Cube Pervious Concrete Samples 




Figure 4.11 Fracture Occurring Through the Aggregate 



111 



4.2 Providing Stormwater Management Solutions in Rajkot, India: A Pervious 
Concrete System Demonstration. As published in The International Journal of the 
Constructed Environment (Solis, Durham, Ramaswami, 2012) 
4.2.1 Introduction 

As part of Rajkot' s development plan, the existing situation of the stormwater 
drainage in the city was evaluated. The assessment revealed that Rajkot is still dependent 
on reservoirs and natural courses (nalas) that exist around the city to redirect stormwater 
to the Aji River which is on the west side of the city. However, present natural courses 
are polluted and frequently used as waste streams. Additionally, many of the natural 
courses have been covered by reinforced concrete slabs due to urban development. 
Natural basaltic roads, hard rock, and mineral soils also make it difficult for stormwater 
to seep into the ground, therefore, allowing water to accumulate on the surface with just 
rainfall intensities of 100 mm (4 in.). (Rajkot Municipal Corporation, 2006). Within the 
development plan, the city expressed concerns that the flooding would cause health 
hazards from accumulation of stagnant water and solid waste around the city. Also, there 
had already been road and property damage identified due to lack of proper stormwater 
drainage. 

Solutions for stormwater management in Rajkot include cleaning existing natural 
courses and installing stormwater pipelines and gutters that link to the natural courses. In 
addition to these solutions, the unique pavement technology known as pervious concrete 
can help Rajkot meet stormwater drainage demands while meeting certain growing 
environmental demands. Pervious concrete has been recognized by the EPA as a Best 
Management Practice (BMP) for reducing stormwater runoff, recharging groundwater, 



112 



and reducing pollutant concentrations (Tennis et al., 2004). BMPs are mitigation 
solutions to the adverse impacts of urban development. Pervious concrete has benefits 
that fall into three categories: environmental, economical, and structural. It is considered 
environmentally beneficial because it has the ability to capture stormwater, filter the 
water as it captures it, and depending on the system design, it can replenish the 
groundwater directly, or captured water can be directed towards a city drainage system. 
Capturing of stormwater can also reduce runoff. In comparison to asphalt pavements, 
pervious concrete will absorb less heat. If used around landscaping, pervious concrete can 
possibly provide water and more air to the trees (NRMCA, 2004). The economical 
benefits might include reducing the number of retentions ponds and reducing the need for 
the large capacity of storm sewers. Structural benefits are due to the texture and strength 
of the concrete. A textured surface due to coarse aggregate exposure provides traction for 
drivers. Strengths of pervious concrete can range between 2.76 to 27.5 MPA (400 to 4000 
psi) (Kosmatka, et al., 2002). 
Plan Modifications 

As stated previously the field installation was going to occur at a waste water 
treatment facility. The location of the field installation changed because the project was 
dependent on construction occurring on-site. In other words, Rajkot Municipal 
Corporation and Lakhlani Associates preferred to use materials for the pervious concrete 
from other projects being constructed nearby. The new location was on location where 
an elevated water tank was being constructed. Figure 4.12 shows the second proposed 
site for the pervious concrete. An AutoCAD drawing of the proposed pervious concrete 
system profile was prepared for the site and is presented in Figure C3 in the Appendix. 



113 



_ti 



Possible area 
designated for the 
v? pervious concrete 

-^#•1 i s y stem 




Figure 4.12 Second Proposed Site for the Pervious Concrete System Placement 

Discussion on the second proposed site resulted in a few concerns: 

• Uncertainty in the quality of materials - The second site was proposed due to 
variability in compressive strength attained from the concrete project occurring at 
the original site. Although it was suggested that changes in weather may have 
caused variable material properties for the concrete at the waste water treatment 
site there was concern that similar problems would occur at the second proposed 
site. It is very important that pervious concrete reach adequate strength 

• Uncertainty in the grading of the land for drainage - The second site appeared 
very flat or there was not proper grading for drainage to flow over the pervious 
concrete. Pervious concrete will only work if there is adequate drainage and a 
holding place for the water. 

• Uncertainty whether there was enough time allotted for construction of the 
pervious concrete system - It was not clear whether workers on site would leave 
for a festival during the month of construction of the pervious concrete 



114 



• Uncertainty of whether the pervious concrete system would be replaced within a 
few months for placement of an asphalt pavement - The original design of the 
elevated water tank site included an asphalt pavement road to be installed before 
the monsoon period. It was not clear whether the contractor would allow a long 
term installment of the pervious concrete system. Long term data was necessary 
for the field demonstration of the pervious concrete system in order to gather 
accurate conclusions about the system subjected to field conditions especially 
during the monsoon period. 

Overall, there was the concern that all parties involved would be investing time, labor, 
materials, and cost with some of these uncertainties and concerns not fully being 
resolved. The representatives of the University of Colorado Denver did not want a 
negative experience to result based on these uncertainties and prevent a new technology 
from being adopted. An alternative was presented to Rajkot Municipal Corporation and 
ICLEI-South Asia such that a smaller demonstration project would be completed to show 
the following benefits of the pervious concrete system: 

• Reassurance of the quality of material for pervious concrete, 

• Assurance of adequate strength gain for the pervious concrete, 

• Demonstration of drainage capabilities and hydraulic conductivity 

• Comparison of water quality before and after percolation of simulated stormwater 
through the pervious concrete system 

The smaller demonstration project involved the construction of a small above ground 
pervious concrete system in a large (approx. 208 L [55 gal]) trash can, barrel, or 
container. The container was filled with three layers of material (150mm (6 in) of sand, 



115 



150mm (6 in) of good draining rock, and 150mm (6 in) of pervious concrete). An outlet 
pipe was constructed at the base to allow water to flow out of the demonstration pervious 
concrete system. All parties agreed to the alternative demonstration and the small 
demonstration project was completed at the water tank site where materials and 
equipment were available for use. Based on compressive strength results of pervious 
concrete samples and the potential to improve water quality of storm water Rajkot 
Municipal Corporation and ICLEI-South Asia may consider a field installation of the 
pervious concrete system at a later date. 

In this dissertation, the results of a small pervious concrete pavement (PCP) 
demonstration executed in Rajkot are presented. The main objective of the demonstration 
was to determine the potential of using a pervious concrete system for stormwater 
management in Rajkot. The test results provided insight into whether the materials 
available in Rajkot were suitable to take advantage of the three main benefits of pervious 
concrete (environmental, economical, and structural). 
4.2.2 Materials and Methods 
Preparation of Base and Sub-base 

A comprehensive design of a PCP includes the drainage, base material, and 
finally the pervious concrete. These three design criteria provide a stormwater 
management system with the capabilities of capturing stormwater, filtering the water, and 
providing durability and resistance to loadings. Preparation of the PCP system 
demonstration required the use of a barrel, with the dimensions 0.62 m (Length) x 0.4572 



116 



m (Diameter) (L 24.5in x D 18 in). A hole was made about 38.1 mm (1.5 in) from the 
bottom of each barrel so that a perforated PVC pipe with an outside diameter of about 32 
mm (1.5 in) could be placed in the barrel and through the hole. The perforated pipe had 
the purpose of collecting and draining the water that percolated through the system. 
Figures 4.13a and b illustrate how the perforated pipes were placed in the barrels. One 
end of the perforated pipe was sealed off with electrical tape to allow water to exit only 
one end of the pipe. The pipes were also provided with about an 8% slope to force water 
to exit through the open end of the perforated pipe. The slope was provided by placing a 
76.2 mm (3 in) brick underneath the taped end of the perforated pipe. Layers of 20 mm (~ 
% in) coarse aggregate and fine aggregate were placed in the barrels (Refer to Figures 
4.14a and b). A cloth fiber was used in place of a geotextile fiber between the coarse 
aggregate and sand layers (Refer to Figure 4.15). The cloth fiber was more readily 
available than the geotextile fiber. The fabric had the purpose of allowing water to 
percolate through the various layers but preventing the fine aggregate from clogging the 
layers of coarse aggregate and pervious concrete. Figure 4.16 shows the schematic of 
how thick the layers of aggregate were and their locations in the barrel. After all layers 
had been placed in the barrel, the layers were compacted with a block of wood that was 
available on site. The layers were allowed to settle for two days and then were provided 
additional compaction by pouring at least two 8 L (2 gal) buckets of water into the barrels 
just before placement of the concrete layer. 



117 





(a) (b) 

Figure 4.13 (a) A Perforated Pipe Placed in Barrel (b) Image of Barrel 




(a) (b) 

Figure 4.14 Base and Sub-Base Layers a) Coarse Aggregate Layer b) Fine 
Aggregate Layer 





Figure 4.15 Cloth Fiber used between Coarse and Fine Aggregate Layers 



118 



A 



Cloth fiber {no gee-text ie fiber 

was readily available-) 



0.032 m diameter PVC pipe, 
with appro* 8% slope 




IP 



0,15 m pervious concrete 



I 



T 

0.15m coarse agg . (size 20 mm) 

I 

0.15 m firveagg 



0075m coarse agg. (size 20 mm) 



T 

Figure 4.16 Profile of the Pervious Concrete System Placed in the Barrel 

Batching and Curing the Pervious Concrete System 

The pervious concrete for the small demonstration was batched during an ambient 
temperature between 29.4°C (85°F) and 32.2°C (90°F), which fell below or equaled to the 
maximum recommended batching temperature of 32.2°C (90°F) (CRMCA, 2009.). The 
mixture design was based on 31 1 kg of grade 53 ordinary portland cement per cubic 
meter of concrete (which is equivalent to a Type II cement at about 525 lb/yd ). The 
design water to cement ratio (w/cm) was 0.30. No admixtures were included in the 
design. The design air content was 13%. The expected air content was based on an 
average of measured percentage voids that resulted from pervious concrete mixture 
experiments performed in lab. 6% fine aggregate of total aggregate was also included in 
the mixture design. The mixture design is shown in Table 4.7 and batch quantities are 
presented in Table 4.8 and the mixture. 



119 



Table 4.7 Mixture Proportions for Rajkot 



Mixture 


w/c 


Water, 
kg/m 


Cement, 
kg/m 


(20 mm) 

Coarse 

Aggregate, 

kg/m 


(12 mm) 

Coarse 

Aggregate, 

kg/m 


Total 

Coarse 

Aggregate, 

kg/m 


Fine 

Aggregate, 

kg/m 


Air 
Content, % 


R 


0.3 


93 


311 


1007 


714 


1721 


111 


13 


1 kg/m' 


= 1.685 lb/yd J 















Table 4.8 Batch Quantities 








Material 


Quantity 


Grade 53 Cement 






26.1 kg (57.5 lb) 


Water/cement ratio 






0.3 


Coarse aggregate 20 mm (0.8 in) 






83.4 kg (183.9 lb) 


Coarse aggregate 10 mm- 12 mm 


(0.4 


in to 0.47in ) 


59.2 kg (130.5 lb) 


Fine Aggregate 






9.2 kg (20.2 lb) 


Water 






9.4 L (2.38 gal) 



The specific gravity of the cement and aggregate was provided from a representative of 

Ambuja Cements Ltd. The values of specific gravity are given in Table 4.9. 

Unfortunately, no one could provide the absorption capacity for the aggregates. It was 

assumed that the absorption capacity would be 1.00 just for simplicity. Additionally, the 

moisture content of the aggregate is either not determined or not frequently determined 

for on-site construction. Therefore, the moisture content was assumed to be zero if the 

aggregate was exposed to the sun and dry weather conditions. 

Table 4.9 Specific Gravity Values Provided used in the Pervious Concrete Mixture 
Design 

Material Specific Gravity 
Cement 3.15 

Coarse Aggregate 2.70 

Fine Aggregate 2.74 

During the process of mixing the batch, the consistency was checked about three times. 

Only about 7.1 L (1.9 gal) of water had been added before the initial assessment of the 

batch. At that time, the mix was too wet. Approximately 0.4 kg (0.9 lb) of cement and 1 

L (0.3 gal) of water was added to the mixer. A second assessment of the consistency was 



120 



made, once again the mix appeared to be too wet. Another 0.4 kg (0.9 lb) of cement and 
0.7 L (0.2 gal) of water was added to the mix. The additional cement and the amount of 
water finally added changed the water/cement ratio from 0.3 to about 0.33. Using Tennis, 
Leming and Akers' (2004) method for consistency identification the final inspection of 
the pervious concrete made in Rajkot was identified as having a "proper amount of 
water" and being "too wet." There was some cement paste left on the hand when the 
pervious concrete was shaped into a ball, and some aggregate stayed intact with each 
other (See Figure 4.17). 




Figure 4.17 Evaluation of Pervious Concrete Consistency 

As stated previously, the base and sub-base materials in the barrel were compacted by 
saturating the material with two 8 L (2.1 gal) of water before placing a 0.15 m (6 in) layer 
of pervious concrete as the final layer in the barrel. Six 15 cm (6 in) cubes (See Figure 
4.18) were made such that three cubes were reserved for 7-day and 28-day compressive 
strength tests. The cubes were made following Indian Standards IS 516 (2002). During 
Phase I laboratory testing cubes were made following the steps shown under section 4.1.2 
During Phase I testing Indian were not available. However, the only difference between 
steps previously discussed in 4.1.2 compared to the Indian Standards was the number of 



121 



roddings used for each lift. Therefore each layer or lift was rodded 35 times versus 25 
times. 

Additional modifications were made when compacting and curing the cubes. For 
example, a wood block (See Figures 4.19a and 4.19b) or steel mold was the tool used in 
the compaction process for the cubes and barrel respectively because a roller or mallet 
was not readily available. 6 mil plastic was also not available, so curing proceeded with 
the use of polypropylene cement bags and wet jute bags. After about 3 or 4 hours of 
initiating the pervious concrete cubes and small demonstration barrel all concrete was 
covered with a wet jute bag (See Figure 4.20). 




Figure 4.18 Rodding the Layers of Pervious Concrete in the Cube Mold 




(a) (b) 

Figure 4.19 Compacting the Pervious Concrete in the Cube Molds Using (a) 
Direction 1 and (b) Direction 2 



122 




Figure 4.20 Covering the Pervious Concrete with a Wet Jute Bag 

Removal of Pervious Concrete from Molds 

The pervious concrete cubes were removed from the molds after curing for 24 
hours. In Phase I lab testing the cubes were cured such that the samples remained in the 
molds and covered in 6 mil plastic for 14 days. Each day the samples were sprayed with 
water. Since the pervious concrete demonstration was occurring on location where the 
water tank was being constructed, the molds were needed for the sampling during the 
water tank construction. While removing the cubes from the molds, the cubes remained 
well intact, demonstrating good cohesiveness (See Figures 4.21a and 4.21b). As indicated 
by IS 516 and ASTM C39, the cubes were placed in a water bath for curing. The concrete 
cubes were placed in empty cement bags and then placed in the water bath which was on- 
site (See Figure 4.22). 





(a) (b) 

Figure 4.21 Removal of Pervious Concrete from Cube Molds (a) Close-Up View 
(b) All Six Cubes 



123 




Figure 4.22 Placing Pervious Concrete Cubes in a Water Bath 
4.2.3 Test Methods and Results 

Percentage Voids and Hydraulic Conductivity Test 

The percentage voids test was completed when the cubes were tested for 7-day 
compressive strength. The procedure used in determining percentage voids was the same 
used in Phase I laboratory testing. Figure 4.23 provides a depiction of how each concrete 
cube was placed in separate containers to determine percentage voids. 




Figure 4.23 Placement of the Pervious Concrete Samples in Water Filled Container 
to Determine Percentage Voids from Volume of Displaced Water 

The average of percentage voids fell within the suggested range (15% to 25%) by Tennis 

et. al (2004) as seen in Table 4.23. Voids are not uniform in size and are not distributed 



124 



evenly within sample, thus, the duration for soaking the samples in water can affect the 
calculations for percentage voids. 

Table 4.10 Results o f the Calculated Percentage Voids 

Sample % Voids 

1 16.9% 

2 18.2% 

3 17.7% 
Average 17.6% 

Two tests were performed to determine a hydraulic conductivity value for the pervious 
concrete and the pervious concrete system separately. Hydraulic conductivity can be used 
to describe the movement of water through the media over time. The hydraulic 
conductivity test for the pervious concrete was performed using the method described in 
section 4.1.2 also known as Delatte, Mrkajic, and Miller's (2007) method of drain time 
test. 

The hydraulic conductivity test for the entire pervious concrete system includes 
the effect of the layers of aggregate, geotextile fiber, and pervious concrete on the 
system. Thus, the hydraulic conductivity test for the system was conducted such that the 
system within the barrel was filled with water until approximately 10.2 cm (4 in) of water 
covered the surface of the pervious concrete. The water was allowed to percolate through 
the system until 7.6 cm (3 in) had drained from the initial height of water. The test was 
performed twice and the drain time was used to calculate the hydraulic conductivity. 
Table 4.11 shows average hydraulic conductivities for the pervious concrete and the 
system separately. Note: Table 4.11 was modified to better represent a falling head test 
and take into account the length and surface area of the pervious concrete and the 
system. Therefore this table differs from that reported in the article by (Solis, Durham 
and Ramaswami, 2012) 

125 



Table 4.11 Hydraulic Conductivity of the Pervious Concrete and System 



Media 


Drain 
Time (s) 


Drain 
Time (s) 


Drain 
Time (s) 


Average Hydraulic 
Conductivity (in/s) 


Average Hydraulic 
Conductivity (mm/s) 


Pervious Concrete 
Pervious Concrete System 


11.5 

1800 


8.5 
563 


9.6 


0.13 
0.012 


3.37 
0.29 



According to Bear (1972), the hydraulic conductivity range for a pervious 
material is 0.1 to 10 cm/s. The pervious concrete had a hydraulic conductivity of about 
0.33 cm/s (0.13 in/s) and thus fell within this range. The hydraulic conductivity for the 
pervious concrete may have been over estimated. This is suggested because the drain 
time for the cylinder mold alone is about 9 seconds. While the drain time for the 
pervious concrete in Rajkot was measured to be an average of about 10 seconds. This 
may have been due to the seal between the hydraulic conductivity apparatus (i.e. cylinder 
mold) and top surface of the concrete being loose. However, this measured hydraulic 
conductivity could be correct because the surface area of the sample is much larger 
compared to the samples tested in lab. The calculation for hydraulic conductivity 
incorporates the surface area of the sample as compared to the surface area of the 
cylinder mold. Using the empirical equation, established by Delatte, Mrkajic, and 
Miller's (2009), for the pervious concrete alone, then the equivalent hydraulic 
conductivity is 9.7 mm/s (0.38 in/s). The empirical equation reports a hydraulic 
conductivity 6.3 mm/s (0.25 in/s) higher than the falling head equation. This difference 
might be explained by the characteristics of the samples tested by Delatte, Mrkajic, and 
Miller (2009). Aggregate gradation is not discussed in their study, however, it was 
mentioned that many of the samples were taken from pavements that were raveling and 



126 



had some clogging. For this study, relating the drain time test directly to the falling head 
equation is preferred. 

The pervious concrete system had a hydraulic conductivity of about 0.03 cm/s 
(0.01 in/s) and fell within a range of 10" to 10" cm/s, which is representative of well- 
sorted sand or a mix of sand and gravel. The pervious concrete system consists of the 
layers of coarse aggregate and sand along with the pervious concrete as the top layer. 
Thus, the controlling layer for the hydraulic conductivity of the system would be 
dependent on the sand, which is represented by the 0.3 mm/s in Table 4.1 1. 
Compressive Strength Testing of Pervious Concrete Cubes 

Three cubes were tested as recommended by IS 516 and ASTM C39 for 
compressive strength at 7 and 28 days of curing (See Figure 4.24). At 7 days, visual 
observations of the tested sample revealed that the cement paste was soft and was still in 
the process of curing. Figures 4.25a and 4.25b show the cement paste had not hardened 
completely (which was expected since maturity can vary based on design of the concrete 
and must be determined by laboratory testing [CRMCA, 2009]) and broke into small sand 
like pieces rather than large stiff chunks. In fact, Phase I testing required that the samples 
remain covered with the 6 mil plastic for 14 days. However, this type of curing was not 
an option for the samples in Rajkot, instead the samples, remained in the curing bath up 
to 7 and 28 days of compressive strength testing. 

The results of the three tests and the average of the three tests for each day are 
reported in Table 4.12. As stated previously, the compressive strength of pervious 
concrete can vary. Traditional concrete pavements can have compressive strengths 
between 20.7 MPa and 34.5 MPa (3000 psi and 5000 psi). It was expected to achieve at 



127 



least a 13.8 MPa (2000 psi) by 28 days. By the 7 1 day, the pervious concrete specimens 
had reached half the strength that was expected (Refer to Table 4.12). However, at 28 
days of age, the sample strengths varied between 5.5 MPa and 13.2 MPa (795 psi and 
1908 psi). Strength results fell within the possible range of applicable pervious concrete 
strengths but were about 4.6% below design strength, which suggests that this particular 
mix design could serve, better, as a pavement for lighter loads experienced by sidewalks. 




Figure 4.24 Compressive Strength Test and Fracture Path 





(a) (b) 

Figure 4.25 Visual Observations (a) The Sample after Completion of Compressive 
Strength Test (b) Breaking the Sample Further by Hand 



128 



Table 4.12 Results of Compressive Strength of Pervious Concrete Samples 





Compressive 


Strength ( 


rf Cubes 






7 day strength 


28 day strength 


Sample 


MPa 


psi 


MPa 


psi 


1 

2 
3 


4.5 

6.2 
8.8 


653.6 

895.4 
1277.6 


7.0 

5.5 
13.2 


1019.1 
795.5 
1908.1 


Average 


6.5 


942.2 


8.6 


1240.9 



Water Quality Testing and Results 

Water from a bore well and water from a nala (stream) on the North-West side of 
Rajkot city were used to test the potential of the pervious concrete system to filter the 
water. Water samples were submitted to the K.C.T. Consultancy Services in Ahmedabad 
for metals testing and to the Gujarat Pollution Control Board in Rajkot for pathogens and 
other property testing. Two sources of water were chosen for the purpose of examining 
the effect that the pervious concrete system had on the water quality of an assumed clean 
source of water (i.e. bore well) versus a source of water that could represent stormwater 
(i.e. stream). 

The water samples were placed in 5 L (1.3 gal) rinsed plastic containers (Figures 
4.26a and 4.26b) and 300 ml sanitized glass bottles (Figures 4.27a and 4.27b). 
Observations of the water samples were made before and after percolation. Figure 4.26a 
shows the well water samples before (container A) and after (container B) percolation 
through the pervious concrete system. The color of the water may have changed after 
percolation through the system as seen with container B. In Figure 4.26b container Y 
holds the stream water before percolation, and container X is the sample after 
percolation. The color of the stream water after percolation is significantly different and 
supports the idea that certain constituents of the water are being filtered. Figure 4.27a and 



129 



4.27b also show the difference in color that occurred with the well and stream water after 
percolation. 




(a) (b) 

Figure 4.26 Before and after Percolation (a) Bore Water Samples (b) Stream Water 
Samples 





(a) 0) 

Figure 4.27 Samples Collected for Pathogen and B.O.D. Tests (a) Bore Well Water 
Samples (b) Stream Water Samples 

Water quality test results for the two sources of water are shown in Table 4. 13. The 

results are compared to available drinking water criteria from BIS IS 10500, Central 

Pollution Control Board in India, and U.S. Environmental Protection Agency (EPA). A 

comparison is also made with the limits for freshwater criteria available through the U.S. 

EPA. Stormwater quality in the U.S. is often compared to freshwater (or individual cities 

or states have established criteria), thus, Table 4.13 uses both drinking and freshwater 

criteria comparisons since freshwater criteria for India could not be found at this time. 



130 



Table 4.13 Water Quality Analysis of the Water from a Bore Well and Stream 



Parameter (units) 


Bore well 


Stream 


Drinking Water 
Limits 


DrinkingWater 

Limits iv o 

conventional 

treatment 
followed by 
disenfection 


Drinking 
Water 
Limits 


Fresh 
Water 
Limits 


Before 


After 


Before 


After 


BIS IS 10500 


Pollution 

Control Board 

India 


EPA 


EPA 


PH 


7.94 




7 63 


6.5-8.5 


6.5-S.5 


5-9 


6.5-9 


Suspended Solids [mg'l) 


6 


12 


68 


36 


NL 


NL 


NL 


NL 


Ammonica. Nitrogen [ms '.) 


028 


028 


8.96 


NL 


NL 


NL 


10 


Iota; Kieidah: Nitrogen {mg 1) 


031 


0.36 


3 4.1 


10.11 


NL 


NL 


NL 


NL 


Nitrite (mg-'i) 


0.0013 


0.507 


0.026 


0289 


NL 


NL 


1 


NL 


Nitrate (mg'l) 


225 


1.13 


- 


- 


45 


NL 


10 


NL 


Total Hardness (mgT) 


230 


130 


380 


180 


200 


NL 


NL 


20-5000 


Total Conform (MPN 100 ml) 


7 


6 







<5Q 





NL 


Peca; Coliform (MPN 100 ml) 


4 


4 




o 


n 





< 126 


Chemieai Oxygen Demand 
(mg'l) 


7 


5 


143 


23 


NL 


NL 


NL 


NL 


BOD (mgT) 


0.6 


- ■ 


NL 


<- 2 


NL 


NL 


Conductivity fyilho cm) 


542 


433 


1260 


553 


NL 


max 2250* 


NL 


NL 


Turbidity (NTU) 


NT 


NT 


0.7 


2.1 


5 


NL 


03 


NL 


Iron (mgT) 


NT 


NT 


0.05 


1.5 


03 


NL 


03 


1 


Chloride (mg'l) 


NT 


NT 


146 


173 


250 


NL 


250 


860 


Flouride (mg'l) 


NT 


NT 


0.75 


0.35 


1 


NL 


4 


NL 


Dissolved Solids (mg'l) 


NT 


NT 


500 


NL 


500 


NL 


Calcium (mg'l) 


NT 


NT 


53.26 


43.56 


75 


NL 


NL 


NL 


Magnesium (mg'l) 


NT 


NT 


22.1 


16.6 


30 


NL 


NL 


NL 


Sulphate (mg'l) 


NT 


NT 


69.81 


75.15 


200 


NL 


250 


NL 


Zinc (mg'l) 


NT 


NT 


035 


03 


5 


NL 


5 


0.12 


Alkalinitv (mg '.) 


NT 


NT 





141 


200 


NL 


NL 


20 


Aluminum (mg'l) 


NT 


NT 


0.06 


0.1 


0.03 


NL 


0.05-02 


O.087 


Free Dissolved CO : (mg'l) 


NT 


NT 


123 


1.92 


NL 


NL 


NL 


NL 


Dissolved 3 {ml) 


NT 


NT 


5.95 


6.55 


NL 


>6 


NL 


NL 



NT - Not tested, NL - Not listed 

*limit based on irrigation, industrial cooling, control of waste disposal 

In Table 4.13, the cells that have been highlighted in red indicate that the results 

are above the standard limits. Also, some values increased after percolation through the 

pervious concrete system. For example, the pH increased in alkalinity for both the well 

and stream water samples. Concrete has a high pH due to the presence of calcium 

hydroxide, which forms from the reaction of Portland cement and water. The findings on 

pH levels are similar to studies performed by Hager (2009) and Caulkins, Kney, 

Suleiman, and Weidner (2010), where pH levels of water, after passing through their 

pervious concrete samples, resulted in pH values between 1 1 and 12. Although, more 

131 



studies are needed to determine whether pH levels reach a neutral level after several 
flushes, it has been suggested that high pH levels could act as good buffers to treat acidic 
water or acid rain (Hager, 2009 and Majersky, 2008). 

Table 4.13 shows turbidity, chloride, and sulphate levels increased after 
percolation but remained below standard limits. Stream water had high values for 
ammonical nitrogen (before percolation), such that they were above EPA's fresh water 
limits. High levels of nitrogen, nitrate, and nitrite might be explained from the leaching of 
human and animal into the stream water. If true, the leaching could be occurring during 
the flooding events. These high levels may also suggest the presence of pesticides and 
inorganic and organic compounds that can cause health problems. But the ammonical 
nitrogen levels decreased after filtration through the pervious concrete, thus, meeting 
EPA's fresh water criteria and suggesting that pervious concrete can beneficially filter 
out high levels of nitrogen. Total and fecal coliforms levels were high before and after 
percolation through the pervious concrete system. A specific value was not determined 
for the coliforms because it was possible that during testing for pathogens, using the Most 
Probable Number (MPN) method, not enough dilutions were made to make a more 
accurate estimate of pathogens. Nevertheless, the presence of coliforms in concentrations 
of 200 MPN/ 100 ml can be associated with disease causing illnesses or organisms that 
are most likely present in the water. In India animals such as cows are allowed to roam 
the streets and any fecal left from the animal could collect in runoff during the rain 
events. This dissertation supports the idea that pervious concrete can filter out some 
pathogens. Although the water quality testing in this dissertation did not support this idea 
(mainly due to a lab error) a study by Luck, Workman, Coyne, and Higgins (2008) 



132 



simulated rainfall over pervious concrete that had been exposed to manure. The effluent 
passing through the pervious concrete was tested for dissolved organic carbon, 
ammonium, nitrate, nitrite, total nitrogen, soluble phosphorus, and fecal coliforms. 
Effluent was tested for three weeks. Within the first week one third of the original 
concentration of fecal coliform forming units (cfu) was detected. By week 2 and 3 fecal 
coliforms were below the detection limits (<2000 cfu/ 100ml) of the device used during 
testing (i.e. spiral plating device). Overall the study indicated that the total reduction in 
coliforms was 10,000 fold compared to the coliforms originally present in the manure. 
The study suggested that the reduction in coliforms might be explained by (1) coliforms 
are trapped in the concrete since some effluent is initially absorbed by the concrete 
material and (2) fecal coliform can die off in alkaline environments. The pH of the 
effluent can exceed 9 due to the concrete as is seen with this dissertation. 

The cells highlighted in yellow also show values that are slightly above some 
standards. These values are no higher than 50% of the highest standard limit such as zinc 
and aluminum. Certain metals such as iron, zinc, and aluminum can have a positive effect 
on human beings. However, exposure to these metals shall meet the minimum drinking 
water level requirements since total intake of such metals already come from other 
sources. Too much exposure to metals can have short and long term health effects. 
Table 4.14 shows results for the majority of the metals that the stream water was tested 
for (before and after percolation). Although Rajkot is known for its manufacturing 
industries (i.e. engines, cutlery, bearings, and casting) surprisingly no metals were 
detected in the samples before and after percolation. If an error occurred during sampling 
before metals testing it may have occurred when nitric acid (HNO3) was directly applied 



133 



to the water-to-be tested. Normally, the containers that will hold the sample are pre- 

cleaned with nitric acid instead of the nitric acid being directly applied to the water 

sample. Additionally, the samples had to be transported for about 3 hours to Ahmedabad 

so the samples could have been affected by improper transportation. 

Note: Currently, Rajkot does not have criteria or standards for stormwater limits. 

Conclusions 

This study demonstrated that the aggregate and cement materials available in 
Rajkot can be used for the construction of a pervious concrete system. The pervious 
concrete and the system revealed reasonable porosity, hydraulic conductivity, and 
filtering capabilities that can be beneficial with the management of stormwater. 
Table 4.14 Additional Results of Stream Water Quality Tests 



Parameter (units) 


Stream 


Drinking 
Water 
Limits 


Fresh Water 
Limits 


Before 


After 


BIS IS 

10500 


EPA 


Residual Chloride Free (mg/1) 


Nil 


Nil 


0.2 


NL 


Copper 


Nil 


Nil 


0.05 


specific calc. 


Manganese 


Nil 


Nil 


0.1 


NL 


Phenolic Compound 


Nil 


Nil 


0.001 


NL 


Mercury 


Nil 


Nil 


0.001 


0.0014 


Cadmium 


Nil 


Nil 


0.01 


0.002 


Selenium 


Nil 


Nil 


0.01 


0.005 


Arsenic 


Nil 


Nil 


0.05 


0.12 


Cynide 


Nil 


Nil 


0.05 


0.0052 


Lead 


Nil 


Nil 


0.05 


0.0025 


Chormium 


Nil 


Nil 


0.05 


0.011 


Mineral Oil 


Nil 


Nil 


0.01 


NL 


Boron 


Nil Nil 


1 


0.75 



Nil - Not detected, NL - Not listed 

Concerns with strength and filtering of pathogens and attaining consistent results in 
all tests have led to a few recommendations. Such recommendations include the 



following: 



134 



• Additional water quality tests should be performed with the same stream water. 
The range of testing (or dilutions) for coliforms should be increased to guarantee 
specific values for before and after filtering of water sample tests. 

• Another pervious concrete mix should be performed using the same mix design to 
check consistency in strength results. If strength results remain below 13 MPa 
(2000 psi), the mix design can be revised to include materials such as silica fume 
and fibers to increase strength and bond strength between the aggregate and 
cement paste. It is also important to note that cube and cylinder strength tests can 
give varying results and should be compared with each other 

• Current strength results reveal that pervious concrete can be used in pedestrian 
pathways or for landscaping where light loads are expected. Literature has shown 
that successful strength is dependent on proper compaction of pervious concrete. 
Quality assurance in compaction can be related to a unit weight test such that 
acceptable values for cylinders range between 1600 kg/m and 2000 kg/m (100 
lb/ft and 125 lb/ft ) (Tennis et. al, 2004). This relation between unit weight, 
compaction, and strength should be investigated further if strength results 
continue to fluctuate. 

• First impressions of the pervious concrete mix and placement suggested that there 
is interest in using the pervious concrete. However, proper field implementation 
requires proper training of employees and appropriate understanding of tools 
needed during placement. Figures 4.28a and 4.28b show the steel roller that might 
have been used for compaction if a large-scale field demonstration of the PCP 
system had been performed. According to studies by Kevern et. al (2009), a steel 



135 



roller should be of appropriate weight. Weights can range between 18 and 30 
kg/m (12 and 20 lb/ft) depending on the workability of the pervious concrete. 




(a) (b) 

Figure 4.28 Steel Roller for Compaction (a) Side View (b) Front View 

4.3 Laboratory Phase II Testing (Cubes Versus Cylinders) 

Throughout this section the author makes reference to results and experiences with Phase 
I testing and Rajkot pervious concrete testing. The goal of the pervious concrete project 
in Rajkot was to demonstrate to the city that future stormwater infrastructure projects 
could incorporate a type concrete technology that also provided environmental, 
economical, and structural benefits. Additionally the demonstration aim to prove that 
Rajkot materials, although different from materials used successfully in the U.S., would 
still contribute to a successful mixture design. In fact the mixture design was adequate 
for achieving common percentages for porosity, hydraulic conductivity fell within a 
range representing pervious material, and there were water quality improvements that in 
categories that are linked to serious health concerns such total nitrogen. 



136 



Additionally, a cost analysis was provided in the article by Solis, Durham, and 
Ramaswami (2012) that showed a reduction in cost spent on materials alone if pervious 
concrete is used in infrastructure projects (approximately 803RS [$17.81] per cubic meter 
[614RS/yd or $13.62/yd ]). However, compressive strength results were variable. The 
standard deviation by 28 days was about 4 MPa (588 psi). At this stage in the project a 
field installation could not be recommended because the design strength was not 
achieved. Although, the strength did fall within the possible strength range (3.5 MPa to 
28 MPa [500 psi to 4000 psi]) indicated in the literature by Tennis, Leming, and Akers 
(2004). 

There is currently no literature that compares the compressive strength of 
pervious concrete cylinders and cubes. Since no cylinders were tested in Rajkot it would 
be beneficial to determine whether the cubes had reached design compressive strength at 
least by applying a factor relating the compressive strength to cylinders. This is 
important, since testing standard requirements use different geometries of specimens. 
Propagation of fractures and types of failures also become important when different 
geometries are tested under compressive strength (del Viso, Carmona, Ruiz, 2007). In 
Phase II laboratory testing it was desired to develop a compressive strength relationship 
between cubes and cylinders. In phase I testing a relationship was not established when it 
was realized that the cube strengths oddly resulted in smaller values compared to the 
cylinders. However, it is good to note that within Mixture 1 the cube compressive 
strength standard deviation was very close to that calculated for Rajkot cubes (what will 
be called Mixture R). Phase II is discussed within the next section. 



137 



4.3.1 Batching and Curing Phase II Laboratory Samples 

The author decided that two more pervious concrete mixtures would be sufficient 
in understanding the relationship between cubes and cylinders for compressive strength, 
porosity, and hydraulic conductivity however, with the main focus on compressive 
strength. Additionally the testing between cubes and cylinders would help identify why a 
high standard deviation in strengths for cubes was occurring (i.e. was it due to batching 
errors, shape of the sample, etc?). Overall the curing process involved removing the 
samples from the molds after 1 day of curing, similar to the samples made in Rajkot. The 
samples remained in a water bath until the day of testing. Samples were tested at 7 and 
28 days of curing. 

Table 4.15 shows the mixture proportioning for Mixture 3 (M3) and Mixture 4 
(M4) and summarizes all mixture proportioning for the previous mixtures. It was based 
on the same mixture design as M2 and MR. The main differences in the design arise from 
specific gravity and absorption capacities. Table 4.16 lists the specific gravities and 
absorption capacities of the material. Similar to Phase I and Rajkot testing the design 
strength was 13.8 MPa (2000 psi). 
Table 4.15 Mixture Proportions for Phase II Laboratory Testing 



Mixture 


W/C 


Water, 
kg/m 


Cement, 
kg/m 


Coarse 
Aggregate, 

kg/m 


Fine 
Aggregate, 

kg/m 


Air 
Content, 

% 


1 


0.3 


93 


311 


1946 


110 


2 


2 


0.3 


93 


311 


1662 


106 


13 


3 


0.3 


93.4 


311.5 


1662.4 


99.4 


13 


4 


0.3 


93.4 


311.5 


1662.4 


99.4 


13 


R 


0.3 


93 


311 


1721 


111 


13 



138 



Table 4.16 Specific Gravities and Absorption C apacities in Phase II Testing 

• , ^ ■<- ^ Absorption 
Material Specific Gravity 
Capacity 



Cement 


3.15 


Coarse Aggregate 


2.60 


Fine Aggregate 


2.64 



0.7 
1.0 



Constructing the small demonstration of a pervious concrete system helped to 
visualize the process of batching concrete in Rajkot, India. For example the aggregates 
are separated in piles by sizes. Therefore, the 20 mm (0.79 in), 12 mm (0.47 in), and 10 
mm (0.39 in) coarse aggregate are in separate piles. The fine aggregate is in a separate 
pile as well. Sieve analysis equipment is not available on-site. Additionally, sometimes 
only the 10 mm (0.39 in) or 12 mm (0.47 in) aggregate is available on-site. For the 
pervious concrete demonstration 20 mm (0.79 in), 12 mm (0.47 in), and fine aggregate 
were available. In order to use an appropriate proportion of 20 mm (0.79 in) and 12 mm 
(0.47 in) aggregate a mixture design was found for a road project in Rajkot. The road 
project used about 59% 20 mm aggregate of total coarse aggregate and 41% 10 mm (0.39 
in) aggregate of total coarse aggregate (CII, NRC, and Ambuja Cements, 2004). These 
same percentages were used for the pervious concrete project and the 10 mm (0.39 in) 
aggregate was replaced with the 12 mm (0.47 in) aggregate. A sieve analysis was 
simulated for the aggregate used in the pervious concrete since a sieve analysis was 
provided for the road project reference. Figures 4.29 a through d compare a sieve 
analysis between the aggregate used in Phase II and the pervious concrete demonstration 
in Rajkot. 



139 



100 < 

90 - 

- 80 - 

£ 70 - 

BO 

1 6 °- 

£ 50 - 

1 40 - 

t 30- 

20 - 

10 - 
















































11 












































































































































































































































































































































































10.00 


1.00 0.10 
Grain size (mm) 


0. 



100 

9(1 



V 








\ 








\ 








V 








4 








4 








q 








H 








'i 








\ 


■« » »-LL- 







10.00 1.00 0.10 0.01 

Grain Size (mm) 



(a) 



♦ 



10.00 1.00 0.10 

Grain size (mm) 




(c) (d) 

lin = 25.4 mm 

Figure 4.29 Sieve Analysis (a) Phase II Coarse Aggregate, (b) Rajkot Coarse 
Aggregate, (c) Phase II Fine Aggregate, (d) Rajkot Fine Aggregate 

From the sieve analyses in Figures 4.29 and 4.29 b, Rajkot aggregate does not 

contain as many fines as the aggregate does in Phase II testing. The fine aggregate 

gradation in Rajkot is comparable to Phase II testing except the size of the fines in Rajkot 

may be slightly larger than that in Phase II testing. Figure 4.30 shows the difference in 

shape between the coarse aggregate available in Rajkot and available in Phase II testing. 

Rajkot' s aggregates are more angular than those used in Phase II. In fact Phase II 

aggregate has a mix of angular and rounded aggregate. According to Tennis, Leming and 

Akers (2004) pervious concrete with rounded aggregate tend to have greater compressive 

strengths than pervious concrete with irregularly shaped aggregate yet irregularly shaped 

aggregate are still good for achieving desired compressive strengths. 



140 



Note: Phase I testing used the same aggregate as Phase II. There was just a small 
difference in specific gravity and absorption capacity between Phase I and II. 




(a) 
Figure 4.30 Coarse Aggregate (a) Rajkot (b) Phase II 

The shape of the aggregate becomes important in various properties of the pervious 

concrete. Irregularly shaped aggregate can affect the pores of the concrete which in turn 

can affect the compressive strength, porosity, and hydraulic conductivity (Sisavath, Jing, 

and Zimmerman, 2001; and Mahoub, et al., 2009; Neptune, and Putman, 2010). 

Aggregate gradation could be a clue to the performance of a pervious concrete mixture 

design (Neptune and Putman, 2010). According to Neptune and Putman (2010) as 

gradation became well-graded the strength increased but the porosity and permeability 

decreased. 

4.3.2 Sample Shape Effects on the Compressive Strength of Pervious Concrete 

The average compressive strength of the pervious concrete specimens made for 

the small demonstration in Rajkot, India was less (1241 psi [8.6 MPa]) than the design 

strength of 2000 psi (13.8 MPa). From research, it is commonly assumed that the ratio 

between cube and cylinder strengths for conventional concrete is 1.25 (Mindess, Young, 

& Darwin, 2003). The 1.25 ratio applied to the average Rajkot specimen strength at 28 

days results in a cylinder strength equal to 992.8 psi (6.8 MPa). However, the laboratory 



141 



tests performed in Phase I suggested that the 1.25 strength relationship does not apply to 
pervious concrete. From Figure 4.36, Ml and M2 produced cylinder strengths about 48% 
and 57%, respectively, higher than cube strengths. In that case the cube to cylinder 
strength ratios could be 0.52 and 0.43. These ratios suggested that the Rajkot cube 
samples would be 2386 psi (16.5 MPa) and 2886 psi (19.9 MPa) as cylinders strengths. 
In this phase of the study the influence of the shape of the specimens is investigated in 
order to assess whether there is a common relationship between cube and cylinder 
pervious concrete properties. Ultimately a strength factor would help define whether the 
strength of the cubes made in Rajkot, India represented a similar strength tested from 
cylinders made in the U.S. using the same mixture design but having different material 
constituent properties. 
Background on Testing for Compressive Strength on Cubes and Cylinders 

The loaded ends of concrete specimens in a compression test experience friction 
from contact with the platens which introduces a lateral confining pressure near the 
specimen ends. The ends of the specimen will try to laterally expand while the platens 
restrain this expansion due to the platens (usually being steel) having a higher modulus of 
elasticity and Poisson's ratio compared to the specimen. 

Shearing and compression stresses are present over the surface of the specimen as 
a result of friction. With an increase in distance away from the top surface the shearing 
stress will decrease and lateral expansion stresses increase. The specimen will break such 
that the top and bottom of the specimen will form into a cone or pyramid approximately 

— d in height (where d is the lateral dimension of the sample). The cone or pyramid is a 
result of the restraint and can influence the result of true strength. Some research has 

142 



shown that a height to lateral dimension ratio approximately equal to 2 is suitable for 
determining true compressive strength. In this study the (h/d) cy under is equal to 2 while 
the (h/d)cube is 1.0. Generally for h/d values less than 1.5, the strength correction factor is 
less than or equal to 0.97; this can also depend on the design strength of the concrete. 
High strength concrete is less affected by h/d ratio. However, compressive strength tests 
on low strength concrete and h/d ratios less than 2 can over estimate strength (Neville, 
1973). 

A cylinder is loaded such that the direction of force applied is perpendicular to the 
cast layers. However, for a cube, the load is usually applied parallel to the cast layers as 
a result of having to test a plane surface. If the properties of the different layers are not 
the same, a layer with a low modulus of elasticity will be susceptible to deformation 
before any of the other layers. If the platen can change inclination during compressive 
testing then the failure of the cube specimen can occur when it reaches the strength of the 
weaker layer (Neville, 1973). In this particular study this is important to keep in mind 
because the testing machine used allowed for the platen to change inclination during 
testing. Also it is critical to note that the information on compressive strength of 
cylinders and cubes has been a result of research on conventional concrete and not 
pervious concrete 
Fracture patterns 

In Figures 4.31 and 4.32 examples of how the samples fractured, during 
compressive strength testing, are shown. During testing vertical cracking, as was seen in 
Phase I testing, was present in Phase II testing. However, as the samples were removed 



143 



from the compression machine the shape of the final sample was usually an hour glass 
shape for cubes or cone for cylinders. 





(a) (b) 

Figure 4.31 Compressive Strength Fractures for M3 (a) Cubes and (b) Cylinders 




(a) (b) 

Figure 4.32 Compressive Strength Fractures for M4 (a) Cubes and (b) Cylinders 

In some cases the failure of the cylinders had the appearance of curvature (see Figure 

4.32) and this is most likely explained as the fracture path following the cement paste 

bond around the aggregate. Some fracture also occurred through aggregate emphasizing 

a good bond between cement paste and the aggregate (See Figure 4.33). Based on these 

failure patterns and comparing them to ASTM C39 the patterns do not seem out of the 

ordinary to conventional concrete. 



144 




Figure 4.33 Fracture Through Aggregate 
4.3.3 Comparing Compressive Strength Results 

Tables 4.17 and 4.18 show the results of compressive strengths for M3 and M4. 
From Tables 4.17 and 4.18 it is important to note that only M3 samples (both cylinders 
and cubes) reached design strength 13.8 MPa (2000 psi). It was promising that the 
design strength was reached at 7 days of curing. However when tested at 28 days of 
curing the cylinders failed to reach, much less pass the design strength. The cubes 
however, had higher strengths at 28 days with the exception for one M3 sample. M4 
samples did not reach design strength at any of the testing days. Additionally during 
Phase II, the cubes tended to have higher strengths than cylinders. 



Table 4.17 Compressive 


! Strengt 


h Results for M3 




Mixture 3 (MPa) 


Mixture 3 (MPa) 


Sample 


cylinc 


ers 


cubes 




7 -days 


28- days 


7 -days 28- days 


1 


14 


13 


14 17 


2 


10 


10 


9 13 


3 


14 


9 


11 15 


Average 


12 


11 


12 15 


Std. Dev. 


2 


2 


3 2 



145 



Table 4.18 Compressive 


j Strength Results for M4 




Mixture 4 (MPa) 


Mixture 4 (MPa) 


Sample 


cylinders 


cubes 




7- days 


28-days 


7-days 28-days 


1 


7 


12 


11 12 


2 


8 


7 


7 13 


3 


8 


8 


13 


Average 


8 


9 


6 13 


Std. Dev. 


1 


3 


3 1 



To summarize Phase I, Rajkot, and Phase II results, Figure 4.34 shows the 
average compressive strength results for M3 and M4 alongside the average results for 
Ml, M2, and MR. All testing results ranged between 6.8 MPa (1000 psi) and 16.4 MPa 
(2380 psi). Based on the average compressive strength results cube and cylinders take 
turns in having higher strengths and this relationship between cubes and cylinders can be 
seen in Figure 4.35. From Figure 4.35 only at 7-day testing can a linear relationship be 
seen for Ml, M3, and M4. By the final day of testing (i.e. 15, 23, or 28 days) half the 
mixtures showed cubes with higher strength and the other half with cylinders having 
higher strengths. 




— • — Ml_cylinders 

— ■ — M2_cylinders 

— * — M3_cylinders 

• M4_cylinders 

- ♦ -Ml_cubes 

- ■ - M2_cubes 

- * -M3_cubes 

- • - M4_cubes 
MR_cubes 

1 MPa =145.038 psi 



Figure 4.34 Average Compressive Strength of Cylinders and Cube Mixes for 
Pervious Concrete Designed for 2000 psi (13.8 MPa) Strength 



146 



T\C\r\ 






ZjUU 

_2000 - 
t 150 ° " 

e 

«> 1000 

U 500- 
n - 






III 






































♦ Ml (7 day) 

■ M2(7day) 
A M3 (7 day) 
X M4 (7 day) 

♦ Ml (23 day) 

■ M2(15day) 
AM3(28day) 






r¥* 


eher Ci 










































ll 


ib 


i 
























































i, 


















ZMrensin 








































































































-^^ 






























































w/ 












































ZL*r 








































































































































X 




























Similar 






=* 




























▲ 
































■ 














Strengt 


IS 










































































































































































































u 


• 


It 




a 










































lllH 


1 


tryi. iiii.fi 
































Strpn pt 


(j 


















































































I | 


















U ^— XM4(28day) 
500 1000 1500 2000 2500 

Cylinder Strength (psi) 



Figure 4.35 Relationship between Cylinder and Cube Average Compressive 
Strengths 

Table 4.19 summarizes the cylinder to cube compressive strength ratio. The ratios 

emphasize the variation in strength results. The ratios range from 0.71 to 2.34. 

Table 4.19 Cylinder to Cube Strength Ratio Based on Average Compressive 
Strengths 

Average Cylinder to Cube 
Ratio at 28-days 



Mixture 7 -day Final day 

Ml 1.11 1.93 

M2 1.78 2.34 

M3 1.08 0.73 

M4 0.87 0.71 



Based on standard deviations ranging between and 4 MPa (580 psi), 7-day and 
final testing day results, and based on a comparison of cylinder to cube strength ratios it 
was decided that a t-test would not be necessary to try to determine if cylinder and cube 
means (for compressive strengths) are RELIABLY different from each other. 



147 



If a t-test was to be performed, 7 day and 28 day strengths would have to be 
analyzed separately throughout the statistical analysis because strength was expected to 
change between these days. In other words the mean compressive strength was 
considered to be moving and the pattern of whether cylinder or cube strengths being 
higher could change as long as the specimen was allowed to cure. Therefore it would be 
necessary to treat 7 and 28 days as two separate tests. However, from Figure 4.36 the 
number of samples in each batch (M2 through MR) has 3 or less samples per testing day. 
It was desired to increase the number of samples for a t-test analysis so all cylinders and 
cube results were combined for 7-day and final day strength results. Figure 4.37 shows 
the average compressive strength results for all cylinders and cubes at 7-day and final day 
compressive strength testing. 




n=3forM3,M4,MR 



n=2 for M2 



I Average Cyl 
I Average Cubes 
Design Strength 



^ sb ^ oft ^ nb ^ rfls 
Batch Codes 



Figure 4.36 Average Compressive Strength with Standard Deviations for All 
Batches 



148 



A 18 

§> 16 
5 14 

CO 

£ 12 
'1 1? 10 

P Oh 

5 w 

a 6 

6 4 

5 - 







1 MPa = 145.04 psi 





I Cylinders 
I Cubes 



7 -day 



Final-day 



Testing Day 



Figure 4.37 Average Compressive Strength with Standard Deviations between 
Cylinders and Cubes at 7-day and Final-Day Testing for all Batches 

4.3.4 Discussion of Standard Deviations and Population 

In order to understand the influence of the standard deviations, the standard 
deviations calculated from the data was compared to other pervious concrete studies. At 
the beginning of this chapter it was indicated that the mixture designs would be based on 
the mixture design by Hager (2009). Hager's study briefly presented standard deviation 
data for compressive strength results. Standard deviation was reported for three types of 
mixtures with ordinary portland cement (OPC), OPC and air entraining admixture, and 
for various mixtures with different percentages of fly ash. This study compares Hager's 
OPC results with the results of this study. Figure 4.37 provides a summary of average 
compressive strength results for all pervious concrete batches minus Ml due to Ml 's 
differing estimated air content. 



149 



It was determined, from Figure 4.37 that there was no statistically significant 
results showing that cube or cylinder strength would be different. In other words, Figure 
4.37 shows that the standard deviations overlap, therefore a t-test is not necessary to 
determine significant differences between cylinders and cubes. Hager's work showed a 
1.3 MPa (188.5 psi) standard deviation at 56 days of curing for 100% OPC concrete 
cylinders. For this study, all cylinders combined have a maximum standard deviation of 
about 2.6 MPa (377 psi) for 7-day testing and 3.6 MPa (527 psi) for final day testing. 
Cubes have standard deviations equal to 2.9 MPa (422 psi) at 7-day testing and 3.8 MPa 
(561 psi) for final day testing. That means this study provides a minimum standard 
deviation difference of about 1.3 MPa (188.5 psi) compared to Hager's work (2009). 
However, it is also important to keep in mind that Hager used a uniform size of aggregate 
in the batching which could help reduce standard deviation. Based on this data and the 
illustrated standard deviations it appears as though more samples should be tested to 
determine if the standard deviations can be reduced. 
4.3.5 Summary of Percent Porosity 

Figure 4.38 shows the summary of percent porosity determined for all batches. 
The percent porosity reveals that at least one percent porosity test from each batch fell 
within the typical range reported Tennis, Leming, and Akers (2004). Except for one 
batch (M3) the average percent porosity falls below the range. 



150 




Typical Range for 
% Porosity 
15 to 25% 



• Minimum 
A Maximum 





v: 




« 




K 




vx 


v: 


— 


<D 


— 


<D 


— 


<D 


— 


(U 


(U 


u 


X 


<D 


Xi 


<D 


X 


<D 


X 


X 


-a 





T3 


3 


T3 


3 


-a 


3 


3 


■M 


U 


:N 


U 


:N 


U 


■N 


u 


U 


>~. 




>> 




>> 




>. 






u 




u 




u 




u 






M 


[1 


M 


[2 


M3 


M4 


MR 



Figure 4.38 Summary of Percent Porosity for All Batches 
4.3.6 Summary of Hydraulic Conductivity 

The hydraulic conductivity was measure using Delatte, Mrkajic, and Miller's 
(2009 and 2007) device. As mentioned in a previous section the device was correlated 
with falling head measurements in order to determine an empirical formula for using the 
drain time to calculate hydraulic conductivity. Using the drain times for each sample 
tested from each batch the hydraulic conductivity reveals that from M2 there were two 
instance where the hydraulic conductivity was representative of a moderately impervious 
batch which occurs at about 0.15 cm/s. However the other two hydraulic conductivities 
calculated for M2 revealed that they were much higher than 0.15 cm/s. It is not clear 
how Delatte et al. method takes into account surface area, because the falling head test 
does. Using the falling head test criteria alone demonstration a much different curve 
compared to that shown in Figure 4.39. For example the drain time was determined on 
cube, cylinder, and larger flat surfaces. Figure 4.40 shows the hydraulic conductivity 
calculate using the falling head equation directly. 



151 



3 

u S 



s 



1.2 

1.0 
0.8 
0.6 
0.4 
0.2 
0.0 



+ 


+ 

* 


\ 
* 


* 


▲r 




■ 




* ▲ 

1 1 1 1 


■ 


■ 



10 20 30 

Drain Time (s) 



40 



50 



♦ Ml* 
■ M2* 

M3cube 
x M3 cylinder 
x M4 cube 

M4 cylinder 

MR** 



Figure 4.39 Summary of Hydraulic Conductivity for all Batches 



_9.0 - 

"£ 8.0 
o 
£7.0- 

:l 6.o 

1 5.0 

§ 4.0 

I 3 '° " 

1 2 -° " 

"Ri.o - 

X 
0.0 

( 










X 


V 




♦ Ml* 


V 




■ M2* 


AM3cube 


X M3 cylinder 


X M4 cube 






• M4 cylinder 


+++ ♦ ■ ■ 


♦ * . . 


MR** 


1 1 1 

) 10 20 30 

Drain time (s) 


40 


1 

50 



Figure 4.40 Summary of Hydraulic Conductivity for all Batches Using Falling Head 
Criteria 

4.4 Summary 

The purpose of this study was to determine environmental and structural 
properties of a pervious concrete demonstration. Changes in rain events can become an 
issue for stormwater solutions so floods are a concern for water quality, capacity and 
long-term durability of stormwater designs. The pervious concrete demonstration 
revealed that Rajkot materials made a pervious concrete batch having porosity and 
hydraulic conductivity that either passed or met criteria. The pervious concrete also 

152 



showed the water filtering capabilities and potential for reducing some polluting 
parameters such as nitrogen levels. However, similar to Hager's (2009) study pH levels 
do increase due to the lime present in concrete. Additionally, the long-term performance 
of strength was determined uncertain (on-average). Cubes only met design strength once 
out of 4 batches, cylinders met design strength once out of 3 batches, and through a 
comparison of standard deviations it was realized a strength relationship between cubes 
and cylinders would not be able to be determined and standard deviations between 
strength results should be reduced. In pervious concrete literature gradation becomes 
important for permeability and strength relationships. Although this dissertation does not 
make a strength and permeability relationship it is good to note that research by Neptune 
and Putman (2010) showed that as gradation became less uniform or single sized and 
more well-graded — the strength also increased, whereas the porosity and permeability 
decreased. In this dissertation the pervious concrete had well graded aggregate. In 
research performed by Mahoub, Canler, Rathbone, Robl, and Davis (2009) the pervious 
concrete permeability and strength did not have a direct correlation, however the degree 
of compaction for lab specimens and field pervious concrete slabs are not accurately 
correlated. Their research revealed that the strength of the specimens compacted by 
using a pneumatic (air pressure adjustable) static press correlated well with field cored 
samples. In this dissertation the lab pervious concrete samples were compacted with a 
roller such that at least 2.5 lb/in was applied to the surface of the samples. However, in 
the literature by Mahoub et al. (2009) the pneumatic press applied at least 8 lb/in and the 
cylindrical samples as well as cored samples reached at least 6.9 MPa (1000 psi) by 28 
days of curing. In this dissertation using roller compaction the samples also reach at least 



153 



6.9 MPa (1000 psi) or greater. It is the opinion of the author that a strength relationship 
is necessary for cross-country comparisons of strength since it is unclear which shape is 
more appropriate for representing strength for pervious concrete, no standards exist for 
testing compressive strength of pervious concrete, and there was much variability in 
compressive strength seen in both data presented for cylinders and cubes in this study. 



154 



5. High Volume Fly Ash Concrete for Hot Weather Conditions: Structural and 

Durability Tests 

5.1 Literature Regarding Fly Ash Use in India 

5.1.1 Properties of fly ash 

Published research on the properties of fly ash concrete presumably first appeared 
in a study by R. E. Davis, Carlson, Kelly, and H. E. Davis in 1937 (Federal Highway 
Administration [FHWA], 1999). The study confirmed fly ash as a type of artificial 
pozzolanic material. Thus, fly ash possessed properties similar to volcanic ash, a 
pozzolana (a siliceous or siliceous and aluminous material having little to no cementitious 
value but in the presence of moisture and calcium hydroxide will react to form 
cementitious properties [ACI 116R,1996] ) used in ancient Rome. The study proved that 
fly ash contributed to concrete strength, could potentially replace cement up to 50%, had 
slightly higher later age strengths than ordinary portland cement concrete, exhibited 
greater plastic flow than portland cement concretes under sustained loading, and had a 
lower heat of hydration (Davis et al., 1937). 

Fly ash is a complex heterogeneous material sometimes having two independent 
and/or a union of two main types of phases which make it difficult to characterize fly ash 
(ACI 232. 2R, 1996). 60 to 90 percent of total fly ash mass can be classified as an 
amorphous (glassy) phase and the other fraction of total mass is a crystalline phase. 
ASTM C 618 provides a method for classifying fly ashes through bulk chemical 
composition. A typical chemistry analysis reports percentage of Si02 (silicon dioxide or 
silica), AI2O3 (alumina), Fe2C>3 (ferrous oxide), CaO (calcium oxide), MgO (magnesium 



155 



oxide), Alkalies (Na20 [sodium oxide] equivalent), SO3 (sulfur trioxide) and Ignition 

Loss (unburnt carbon content). If the sum of SiC>2, AI2O3, and Fe203 is greater than 70% 

the fly ash is a Class F and if the sum is less than 70% it is a Class C. Fly ashes usually 

contain some lime thus their cementitious value will vary even without the addition of 

calcium hydroxide from portland cement. Thus from the chemical analysis a Class C fly 

ash will have 20% or more CaO content. There are differences in chemistry for fly ashes 

from country to country. Table 5.1 shows an example of chemical compositions of fly 

ashes from different countries. However, a chemical composition neither addresses 

reactivity nor long-term performance when the fly ash is used in concrete (ACI 232. 2R, 

1996). According to Malhotra and Mehta (2008) the chemical differences are not as 

important as the mineralogical (glassy and crystalline phases) and granulometric (particle 

size and shape) differences. Nevertheless most countries only perform a chemical 

analysis on fly ash. 

Table 5.1 Example of Chemical Composition of Fly Ash from Different Countries 

(Malhotra & Mehta, 2008) 

Percent by Mass 
Constiuent 



Australia India Japan Canada U.S. 

Si02 592 53.0-71.0 544 481) 55T 

A1203 23.4 13.0-35.0 31.1 21.5 21.1 

Fe203 4.1 3.5-12.0 4.6 10.6 5.2 

CaO 2.8 0.6-6.0 4.4 6.7 6.7 

MgO 1.1 0.3-3.2 0.8 1.0 1.6 

Alkalies 0.8 N/A 0.6 1.4 3.0 

S03 0.2 0.01-1.1 0.4 0.5 0.5 

Ignition Loss 1.7 N/A N/A 6.9 0.6 

The glassy phase, which is dependent on the calcium content, highly influences 
the pozzolanic activity of the fly ash. The high calcium fly ashes are more reactive than 
low calcium fly ashes. Easily reactive glass and crystalline minerals include calcium 



156 



aluminosilicate glass, tricalcium aluminate, calcium aluminosulphate, anhydrite, and free 
CaO which is present in the high calcium fly ashes. The high calcium fly ash will have 
cementitious and pozzolanic properties. Low calcium fly ashes have aluminosilicate 
glass which is slightly reactive. The reaction of fly ash with cement mainly depends on 
the breakdown of the glassy phase by hydroxide ions and heat from the early hydration of 
cement. Calcium silicate hydrate (CSH) forms as calcium hydroxide is consumed from 
the reaction of the fly ash. Fly ash particles can range in size between 15 and 20 um. 
The shape of the particles is spherical thus the shape provides a positive effect for the 
workability when used in concrete mixtures. 
5.1.2 Fly ash Consumption in India 

Data regarding fly ash use in India is not easily assessable. Currently, a strong 
organization such as the American Coal Ash Association does not exist in India that 
keeps record of fly ash use. However, some literature has reported fly ash use in India. 
In 2005 112 million tonnes (123 short tons) of fly ash was produced by thermal power 
plants in India (Dhadse, Kumari, Bhagia, 2008). Fly ash benefits are recognized as 
reducing heat of hydration in mass concrete, preventing alkali damage, saving 1 1% in 
cement manufacturing costs, and improvement in concrete strength at later ages for even 
high dosages of fly ash concrete (Dhadse et al., 2008). The consumption of fly ash in 
India has improved since the 1994 levels. In 1994 3% of production was utilized versus 
38% in 2005 (See Table 5.2). Industrial and construction activities in India recognize the 
benefits of using fly ash. As mentioned in Chapter 1 fly ash in concrete is known for its 
benefit in reducing the heat of hydration in comparison to ordinary portland cement 
(OPC). In fact Indian cement industries promote the use of their blended cements (i.e. 



157 



Portland pozzolana cement having fly ash) for reduced heat of hydration. The 

government has made initiatives in promoting fly ash use in bricks, agriculture, and in 

water and waste water treatment plants. Table 5.2 shows the percentage consumption of 

fly ash by certain industries or sectors in India. 

Table 5.2 Year 2005 Production and Utilization of Fly Ash in India 
Sector %* 

Total fly ash produced (million tonnes) 112 



Utilization (million tonnes) 


42 


cement 


49 


roads and embankment 


21 


low lying area filling 


17 


raising dykes 


4 


bricks 


2 


mining 


2 


agriculture. 


1 


Others 


3 



*as a percentage unless noted otherwise 
Source: Dhadse, Kumari, Bhagia, 2008 

Fly ash, in India, is labeled a hazardous material, but the government still 
encourages beneficial use. It is considered hazardous due to its potential to pollute the air 
and water if not properly disposed. In addition, it is consider hazardous because it can 
settle on the leaves of crops that surround power plants thus lowering crop yield. 
Disposal usually requires slurry ash ponds that experience surface runoff during rain 
events so salts and metals leach into the groundwater (ENVIS Centre et al., 2007). 
Disposal costs usually range between Rs 50 -100 ($0.90 to $1.78) per million tonne. 

The cement industry is the largest consumer of fly ash including the hazardous 
material called blast furnace slag. Since the cement industry is a major user of railways, 
the Railway Board in India is considering developing a policy for the movement of fly 
ash which would require investments in infrastructure facilities at loading and unloading 

158 



points (CM A, 2010a). During the 2009 year the Ministry of Environment and Forests 
issued new notifications (No.S.O. 2809) regarding fly ash, that the cement industry 
believed would seriously impact operations (CM A 2010a). Before the notification, fly 
ash was available to end users (i.e. cement industry) without requiring payment except 
for transportation cost. With the new notifications cement manufacturers were no longer 
exempt from having to pay for the use of fly ash. Another clause in the notification 
stated that fly ash must be used if within 100 km (62. 1 mi) radius of a thermal power 
plant. The cement industry felt like this statement restricted the use of blast furnace slag 
which can be used for higher replacements of clinker (up to 65% for slag versus 35% for 
fly ash) (CMA, 2010a). 

Within this study the use of fly ash in buildings and pavements is encouraged for 
cities. Cities such as Rajkot, India, have experimented with high volume fly ash (HVFA) 
concrete and have been successful in attaining good strength as was discussed in Chapter 
2. A local structural engineer, along with the city of Rajkot, had an interest in 
experimenting with other sources of fly ash. In 2004, the HVFA road project included 
the use of fly ash from the Sikka thermal power plant. However, fly ash is also available 
from the Vanakbori and Gandhinagar thermal power plants and it was desired to use 
these fly ashes for this study. The following section presents the experimental study that 
involved the use of two additional sources of fly ash and presents the data regarding fly 
ash properties and compressive strength. 

5.2 Literature on HVFA Concrete for Hot Weather Conditions 
Testing OPC concrete and concrete with cementitious materials, such as fly ash, under 
hot weather conditions has resulted in several published literature. In example Schindler 



159 



(2004) presents a model to account for the effect of temperature on the rate of hydration 
when different cement types and mineral admixtures are used. The model, called 
activation energy model, is a function of the cement composition, and type of mineral 
admixtures used in the entire mixture. The model is created using a nonlinear regression 
analysis. In work by Zhang, Shen, Zhou, and Li (201 1) compressive strength and tensile 
strength development for fly ash concrete samples exposed to thermal environments is 
explained. Both tensile and compressive strength increases with increasing curing 
temperature when concrete samples contain fly ash. Thus their work implied that 
appropriately increasing curing temperature could improve the compressive and tensile 
strength of fly ash concrete but not OPC concrete samples specifically made with a w/c 
ratio of 0.3. However, curing occurred in steamed heated conditions for 6 hours and then 
samples were placed in room temperature for 2 months. In research by Khoury (2006) 
shrinkage, creep, and expansive strains in previous cured plain concrete exposed to heat 
were shown to be dependent upon temperature cycles. However, temperatures ranged 
1 lOoC to 6OO0C to represent nuclear reactor concrete. Nevertheless the paper revealed 
that the long term durability of the concrete is dependent on the aggregate type, the age of 
the maturity of the concrete, and the thermal loading cycles. Within this dissertation, 
however, the literature provided throughout this dissertation pertains specifically to the 
performance of HVFA concrete in heated conditions below 65.6oC (120oF) 

Hot weather is defined as high ambient temperatures (sometimes above 26.7°C 
[80°F]) with any combination of high wind velocity, low relative humidity, and exposure 
to solar radiation (PCA, 2002). Hot weather is detrimental to fresh concrete because it 
can increase water demand, accelerate slump loss, increase the rate of setting, increase 



160 



risk of plastic and thermal cracking, affect entrained air, and increase strength loss over 
long-term due to increases in internal concrete temperatures (PCA, 2002). Hot weather 
effects on hardened concrete have been identified as decreasing strength, decreasing 
durability from cracking, increasing permeability, and increasing risk for drying 
shrinkage (PCA, 2002). Studies have shown the importance of cooling materials before 
mixing, batching and placing concrete during temperatures greater than 85°F and 
publishing these suggested guidelines in books such as the PCA manual (2002) and ACI 
305 (2010) . 

Other studies have subjected concrete samples to hot weather conditions and have 
included usually one of following variables: 

• Humidity in the range of 65% to 100% (Ravina, 198 1 ; Mehta, 2002) 

• Including fly ash at percentages between 30% to 75% (Ravina, 1981; Mehta, 
2002; Senthil & Santhakumar, 2005; Bentz, Peltz, Herrera, Valdez, & Juarez, 
2010). 

• Testing in temperatures up to 40°C (104°F) (Ravina, 1981) 

• Testing with heated materials such as increasing aggregate temperature up to 
70°C (158°F) (Mouret, Bascoul, Escadeillas, 1997) 

Each of the studies that included one of the testing variables above produced information 
about what to expect in terms of compressive strength. The studies indicated that design 
compressive strength could decrease anywhere between 5 and 15%. But, the studies used 
different testing variables under heated conditions, thus it was not clear which of the 
testing variables, including the heated conditions, could be affecting the results of the 
compressive strength more. Additionally, certain studies were unique and should be 



161 



investigated further. For example, Villarreal & Torres (2002) not only subjected their 
concrete samples to heated conditions they cured the samples for a period of 6 months 
under dry conditions and mimicking shading. This study is unique because it might be 
used to represent a climate change effect where extreme temperatures occur over long 
periods of time. However, the samples contained a mix of high volume fly ash (up to 
53% fly ash), and silica fume. So any benefit towards compressive strength may not be 
attributed to the fly ash alone. The study by Mouret, Bascoul, Escadeillas (1997) is one 
the very few studies that tests the effect of heated aggregate to concrete while mixing and 
curing. However, the concrete samples only include ordinary portland cement and the 
samples were initially cured in hot weather conditions for only 24 hours. 

Based on the literature there is a lack of tests on HVFA concrete in semi-arid to 
arid conditions, with heated aggregate, and curing above 37.8°C (100°F). Additionally 
two other testing variables that have not been used in hot weather testing have been 
aggregate content and curing in cyclic temperatures to simulate the thermal gradients that 
arise during the changes between day and night. 
Goal of HVFA concrete in hot weather conditions Study 

Overall the goal of this portion of the dissertation was to evaluate HVFA 
concrete's contribution to climate adaptation in cities. 
Objectives for the Study 

The overall objective to reach this goal was to determine the structural and 
durability benefits that arise from HVFA in concrete mixtures when subjected to hot 
weather conditions. 



162 



This study was unique because it combined many of the testing variables from the 
literature review for a more holistic approach to testing HVFA concrete in hot weather 
conditions. The following testing variables were used: 

• Cement or 50% fly ash 

• Heated or no heated aggregate (Temperatures around 65°C [149°F]) 

• Curing in ideal (water) or dry heated conditions (above 37.8°C [100°F]) 

• Curing for 90 days 

• Changing the aggregate content to 55% coarse aggregate or 65% coarse 
aggregate of total aggregate weight. 

Also the heated curing conditions were set up to simulate cyclic temperatures (or the 
thermal gradient that occurs between night and day). This condition is discussed more in 
detail in the methods section. Also the aggregate content was varied because drying 
shrinkage was identified as a problem in hot weather conditions and has not been 
thoroughly discussed in any of the literature found for this dissertation. PCA (2002) does 
indicate that the shrinkage is best minimized with a low water content. A low water 
content is achieved by using a high coarse aggregate content. Additionally, PCA 
indicated that supplementary cementing materials will have little effect on the shrinkage 
if used in small dosages. Therefore, it was worth exploring the effects of a high dosage 
of fly ash on shrinkage combined with two different aggregate contents. 

Also unique to this study is the combination of testing methods to quantify major 
changes in HVFA concrete samples cured in hot weather conditions. Most literature 
referred to compressive strength to quantify the effects of testing in hot weather 
conditions, but within this study permeability, length change, and modulus of elasticity 



163 



were also measured. Additionally, it was important to this study to test specimens from 
56-days beyond because the age of strength acceptance for HVFA concrete should be 
extended to 56 or 90 days (HR, 2005). 

5.3 Phase I study for HVFA in Hot Weather Conditions: India and U.S. Comparison 
of Fly Ash Properties (Fly Ash Used in Rajkot, Gujarat, India and Denver, 
Colorado, U.S.). 

The tests involving HVFA concrete under extreme hot weather conditions could 
not be performed in Rajkot, India. Instead, two sources of fly ash from thermal power 
plants near Rajkot were tested in HVFA concrete mixtures for compressive strength. 
Rajkot Municipal Corporation and Lakhlani Associates were interested in the 
compressive strength benefits that these two sources of fly ash could potentially provide, 
thus the reason for using these sources of fly ash. The fly ash came from the Vanakbori 
and Gandhinagar power plant located approximately 302 km (187.7 mi) and 258 km 
(160.3 mi) away from Rajkot, respectively. Figure 5.1 shows the two sources of fly ash. 
In Figure 5.1 it appears as though the Vanakbori fly ash has a lighter grey color compared 
to the fly ash from Gandhinagar. The goal for testing these two sources of fly ash was to 
develop a compressive strength relationship with the U.S. source of fly ash. Very similar 
to the method used in Chapter 4 for the pervious concrete, concrete cubes made in Rajkot 
were tested for compressive strength and finally compared to the concrete cubes and 
cylinders made in the U.S. 



164 




Figure 5.1 (a) Vanakbori Fly Ash, (b) Gandhinagar Fly Ash 

In Table 5.3 the chemical analysis for the different fly ash sources are presented. Class C 

fly ash is provided, to show what lime contents levels have to be in order to be classified 

as Class C fly ash. Since the Indian sources of fly ash had very low lime contents their 

reactions with cement would be closer to that of a U.S. Class F fly ash. Class F fly ashes 

have less of a cementitious reaction compared to a Class C fly ash and are therefore 

dependent on the reaction with the lime (calcium hydroxide) in cement to make hydrated 

calcium silicate (which provides strength to the concrete). Additionally, the concrete 

strength gain for Class F fly ash is slower than if Class C fly ash is used. 

Table 5.3 Chemical Analysis for Various Fly Ash Sources between the U.S. and 
India 





India 


India 


India 


U.S. Boral 


U.S. Boral 


ASTM C618 


Chemical 


Sikka 


Vanakbori 


Ghandinagar 


Class F 


Class C 


Class F/C 


Si0 2 (%) 


60.21 


59.00 


58.95 


56.45 


33.64 




A1 2 3 (%) 


26.08 


26.72 


28.49 


21.06 


18.26 




Fe 2 3 (%) 


4.8 


5.42 


4.87 


4.12 


5.27 




Si0 2 +Al 2 3 +Fe 2 3 (%) 


91.09 


91.10 


92.31 


81.63 


57.17 


70.0/50.0 min 


CaO (%) 


1 


3.85 


3.13 


10.83 


27.67 




MgO (%) 


0.3 


0.95 


0.67 


2.29 


6.96 




Total Alkali as Na 2 (%) 


0.86 


0.14 


-0.02 


1.45 


2.27 




so 3 


0.25 


0.16 


0.08 


0.43 


2.18 


5.0 max 


LOI (%) 


1.71 


0.00 


0.00 


0.67 


0.26 


6.0 max 


2 

Fineness (m /kg) 


330 


- 


- 


- 


- 





165 



In Rajkot, two batches of concrete were mixed, with each batch having 50% replacement 
of cement by fly ash. Batching and placing concrete in the molds followed BIS 516. 
One batch contained fly ash from the Vanakbori thermal power plant and the other batch 
contained fly ash from the Gandhinagar thermal power plant. Table 5.4 presents the 
mixture design for both batches. While placing the concrete in the cubes the 
Gandhinagar fly ash exhibited slightly more flowability than the Vanakbori batch. Both 
batches were made using the same amount of materials. The difference in flowability 
may have been due to the differences in fly ash, but may have also occurred from the 
moisture trapped within any of the material since moisture content was not determined 
(i.e. dry material was used but some sand may have been moist just from observation of 
the material out in the field). Or the method of adding water to the batch may have not 
been consistent (i.e. water was added using a 1 liter bottle and any fraction of water that 
was needed was estimated from the 1 liter bottle). Figures 5.2 and 5.3 show the two 
concrete batches before placing into the cube molds and after placement into the molds. 



lable 5.4 


Mixtur 


e rropc 


rtions I 


or HV1 


<A (Joncr 


ete in Kaj 


kot 
















(20 mm) 


(12 mm) 


Total 
















Coarse 


Coarse 


Coarse 


Fine 


Air 






Water, 


Cement, 


Fly Ash, 


Aggregate, 


Aggregate, 


Aggregate, 


Aggregate, 


Content, 


Mixture 


w/c 


kg/m 


kg/m 


kg/m" 


kg/m 


kg/m 


kg/m" 


kg/m" 


% 


Wanakbori 


0.4 


156.6 


195.8 


195.8 


604.1 


428.3 


1032.3 


782.7 


2 


Gandhinagar 


0.4 


156.6 


195.8 


195.8 


604.1 


428.3 


1032.3 


782.7 


2 


1 kg/m 3 = 


1.68554 


!• lb/yd 3 

















Three 15.2 cm (6in) cube specimens from each batch were tested for compressive 
strength after 7-days and 28-days of curing. Figures of the compressive strength and 
fractures paths are shown in Appendix D (Figures D.l through D.2). A 56-day testing 
could not be performed because not enough molds were available on-site where the 



166 



batching occurred. In Table 5.5 the compressive strength results as well as standard 
deviations are shown. Figure 5.4 graphs the compressive strength results. 




Figure 5.2 Batches (a) Vanakbori and (b) Gandhinagar 




Figure 5.3 Cubes (a) Vanakbori and (b) Gandhinagar 

Table 5.5 Compressive Strength Results for Rajkot HVFA Concrete Samples 





Vanakbori (MPa) 


Gandhinagar 


(MPa) 


Sample 




cubes 






cubes 






7- days 




28-days 


7-days 




28-days 


1 


16 




26 


15 




23 


2 


18 




28 


14 




28 


3 


16 




24 


13 




23 


Average 


17 




26 


14 




25 


Std. Dev. 


1 




2 


1 




3 



lMPa= 145.038 psi 



167 



30 



25 



a. 



S20 



S 



55 15 



ft 10 



o 
U 




-Vanakbori 
-Gandhinagar 



28 



lMPa= 145.038 psi 

Figure 5.4 Average Compressive Strength Result for Rajkot HVFA Concrete 

Samples 

The design compressive strength was 27.6 MPa (4000 psi). The compressive 
results reported in Table 5.5 and Figure 5.4 demonstrate that the Vanakbori samples 
gained strength more quickly compared to the Gandhinagar samples. This might be 
explained by the slightly higher lime content that Vanakbori fly ash had according to the 
Table 5.3. By 28-days at least one sample from each batch reached design compressive 
strength. However, average compressive strength results demonstrate show that the 
samples are about 9% less than the design compressive strength. 

The batching of the HVFA concrete samples using the U.S. Class F fly ash from 
Craig thermal power plant followed the same mixture proportions used in Rajkot but 
involved the use of ASTM standards. Table 5.6 shows the mixture proportions. The 
aggregate content for the U.S. mixture differed because specific gravities differed 
compared to those assumed for the aggregate available in Rajkot. Two 15.2 cm (6in) 



168 



cubes and two 10.2 cm x 20.3 cm (4 in x 8 in) cylinders were made for 7, 28, and 56-day 
testing. Slump, unit weight, and air content were measured for the samples batched 
using the U.S. fly ash source (See Table 5.7) while these measurements could not be 
made for the Rajkot HVFA batches because equipment was not available. 
Table 5.6 Mixture Proportions for HVFA Concrete in Denver 



Mixture 



W/C 



Water, 
kg/m 



Cement, 
kg/m 



Fly Ash, 
kg/m 



Coarse 
Aggregate, 

kg/m 



Fine 
Aggregate, 

kg/m 



Air 
Content, 

% 



Craig Class F 0.4 156.6 195.8 



195.8 



996.7 



771.5 



1 kg/m 3 =1.68554 lb/yd 3 

Table 5.7 Fresh Concrete Properties for the HVFA Concrete Batch in Denver 



Craig Class F Fresh Concrete Properties 


Slump 


0.81 in 


20.5 mm 


Unit Weight 


147 to/ft 3 


2354.7 kg/m 3 


Air Content 


1.98% 


- 



The unit weight is very close to a normal weight concrete of 2403 kg/nr (150 
lb/ft ). The slump is low and is mainly due to no use of admixtures however, the 
concrete remained workable throughout the batching and molding process. The 
estimated air content was within 99% of the actual air content. Air entraining admixtures 
were not used in either one of the HVFA batches because high air contents are mainly 
preferable to resist freeze/thaw effects. In Rajkot, freezing/thawing is not a concern. 

Compressive strength results are shown in Table 5.8 and Figure 5.5. Table 5.8 
shows the individual sample compressive strength results. Unlike the Rajkot fly ash 
samples none of the U.S. fly ash samples reached design compressive strength by 28- 
days, however, by 56-days of curing all U.S. fly ash samples passed design strength. 



169 



Table 5.8 Compressive Strength Results for U.S. HVFA Concrete Samples 



Boral (Craig Power Plant) Class F fly ash 


Sample 


Cylinders (MPa) 
7-days 28-days 56-days 


Cubes (MPa) 
7-days 28-days 56-days 


1 

2 
Average 
Std. Dev. 


22 27 30 

21 26 33 

22 26 31 
1 2 


17 22 30 
31 31 

17 27 31 
7 1 



1 MPa = 145.038 psi 




-A-Craig cylinders 
-*-Craig cubes 
-♦-Vanakbori 
-■-Gandhinagar 



lMPa= 145.038 psi 

Figure 5.5 U.S. and India HVFA Concrete Average Compressive Strength Results 

Figure 5.5 also provides a comparison of average compressive strength results for 

the HVFA concrete samples batched in India and in the U.S. By 7-days of curing the 

cylinders gain more strength at a faster rate compared to all cubes. However, the Craig 

(U.S. fly ash) cube strength on average maintains a higher strength value compared to the 

Vanakbori and Gandhinagar samples. Figure 5.6 shows the average compressive strength 

results and standard deviations for all HVFA concrete samples. This figure provides 

another method for extracting observational results between all mixtures. For example, at 



170 



28 days, based on the trend in average compressive strength and standard deviations, 
there appears to be no statistical significant difference among the various sources of fly 
ash compared to the U.S. cylinders. This comparison is necessary to assume that hot 
weather concrete test results gathered from U.S. HVFA concrete would be similar if 
gathered from the HVFA concrete made from the Indian sources of fly ash. Therefore, 
the average cylinder to cube compressive strength ratio reported in Table 5.9 may be 
assumed valid. 



40 








; 



■ Craig 

Li Vanakbori 
tl Gandhinagar 

■ Craig (Cyl) 



7-days 



28-days 56-days 



Figure 5.6 Summary of Average Compressive Strength Results and Standard 
Deviations between the U.S. and Indian Sources of Fly Ash 

Table 5.9 Average Cylinder to Cube Compressive Strength Ratios for U.S. and 
Indian HVFA Concret e Mixtures 

Average Cylinder to 
Cube Ratio at 28-days 
Wanakbori 1.01 

Gandhinagar 1.05 

Craig 0.98 



Gathering the information on U.S. and Indian fly ash relationship revealed that all 
sources of fly ash are representative of a Class F designated fly ash. The strength results 
also showed that HVFA concrete made from these three sources of fly ash produce 

171 



average compressive strength results with only about a 5% difference. Thus tests 
regarding HVFA concrete under extreme hot weather conditions were proceeded with 
using the U.S. Class F fly ash from Craig power plant with the assumption that Rajkot 
HVFA concrete samples would perform similarly with at most a 5% difference. 
5.4 Phase II: Properties of HVFA and OPC Concrete When Subjected to Hot 
Weather Conditions 

To begin testing HVFA and OPC concrete samples in hot weather conditions, 
average aggregate temperatures when exposed to hot weather temperatures were 
measured in order to determine the average temperatures to be used in the laboratory 
testing. Also, the benefit of fly ash concrete having a lower heat of hydration was 
verified through two tests comparing internal temperatures of HVFA and OPC concrete 
samples as they cured in ambient and simulated hot weather conditions. 
5.4.1 Aggregate Temperatures 

The PCA (2002) manual provides guidelines for managing material properly 
during hot weather conditions. Materials should be cooled or protected enough so the 
concrete temperatures can ideally remain around 16 to 27°C (60 to 80°F) (although ACI 
305 does mention the maximum allowable fresh concrete temperature can be 35°C 
(95°F)). As such ready mix companies, at least in the U.S., accomplish this by shading 
with silos and sheds (See Figure 5.7), and spraying or fogging with water. However, 
when spraying, the moisture content of aggregates, before use in a concrete mixture, must 
be taken into account. 



172 




(a) (b) 

Figure 5.7 Aggregate (a) Storing and Cooling in a Shed and (b) Stockpiling 

Within this dissertation the author is suggesting that changes in climate could 
affect the effectiveness of these methods for keeping aggregate cool, especially in 
countries, where, aggregate is most likely stored on-site without shade (sometimes 
referred to as stockpiling [See Figure 5.7 (b)]). Changes in temperatures can require 
more use of water in areas that already experience droughts. The availability of ice is 
already limited in many countries and may not be a priority for concrete. The only option 
for cooling aggregates may be shading, but for regions that already experience hot 
temperatures shading may not help too much once temperatures pass a certain range. 
Additionally, moisture content may not be taken into account on a daily basis. On the 
other hand countries may purposely not cool aggregate by means of water (i.e. dry 
aggregates are preferred and thus it is ok to assume moisture content is negligible). 

Aggregate has the greatest mass in a concrete mixture; taking up 60 to 80% of the 
volume of a normal weight concrete mixture (ACI, 2010) therefore aggregate can have a 
great effect on the temperature of the concrete and final performance of the concrete 
mixture while curing and after curing. In countries where stockpiling is common it is 



173 



possible that there are cases when aggregate is exposed to weather conditions without use 
of shading or cooling by water. Therefore, it was decided that for this research it was 
worth determining what were some of the temperatures aggregate could reach under hot 
conditions [>27°C (80°F)]. For several days from mid-July through end of September 
temperatures of stock piled aggregate were measured and compared to stored or cooled 
aggregates. Stock piled aggregate temperatures were taken on-site a ready-concrete plant 
in Colorado, U.S. (Company Name: Boral Ready Mixed Concrete Company). Originally 
the stock piled aggregate temperatures were going to be recorded on location a 
construction materials company near Phoenix, Arizona (Vulcan Materials Company); 
however, most construction materials companies in hot regions of the U.S. 
unquestionably cool their aggregate. In this case Vulcan Materials Company in Arizona 
did record temperatures of their cooled aggregate stored on-site and their cement and fly 
ash stored in silos. At Vulcan Materials the aggregates are maintained at cool 
temperatures by spraying them and keeping them shaded. The temperatures of the 
aggregate are reported in Figure 5.8. 

From Figure 5.8 the stored/cooled aggregate remains between a range of 20 to 
30°C (68 to 86°F) under ambient temperatures of about 40°C (104°F). Above an ambient 
temperature of 40°C (104°F), it might be assumed that the temperature range of 
stored/cooled aggregate would be difficult to maintain between 20 to 30°C (68 to 86°F). 
The temperatures of the stored/cooled aggregate might even reach 30 to 40°C (86 and 
104°F) in 40°C (104°F) weather. On the other hand, stockpiled aggregates can reach 
temperatures that range between 50 and 75°C (122 and 167°F). Stockpiled temperatures 
can be almost twice the ambient temperatures. These observations made from Figure 5.8 



174 



supports one of the hypotheses of this dissertation (i.e. if ambient conditions can reach 
high enough temperatures it will be difficult to keep aggregates cool and there could be 
more demand on energy and resources to keep the aggregate cool). 



80.0 




32 32 32 35 35 38 
Ambient Temperature (°C) 

-Stockpiled Coarse Aggregate Stockpiled Fine Aggregate 
-Wet Coarse Aggregate Wet Fine Aggregate 



°F = [°C*(9/5)]+32 

Figure 5.8 Temperatures of Stock-Piled and Stored/Cooled Aggregate 

5.4.2 Verifying Temperatures of HVFA and OPC Concrete During Hydration 

A benefit of HVFA concrete is its internal temperature as hydration is occurring 
during curing. The internal temperatures within HVFA concrete can peak about 57% 
(w/c-0.53 and 50% replacement of OPC by FA) less than OPC concrete depending on the 
water cement ratio (Wang and Yan, 2006). In other research the temperature rise in the 
HVFA concrete was about 36% less compared to OPC concrete (Atis, 2002). The lower 
heat of hydration minimizes the risk of cracking which is beneficial during concreting in 
hot weather conditions. For this dissertation it was preferred that these lower internal 
temperatures be verified through some trial mixtures. The mixture proportioning is 

shown in Table 5.10. Trial 1 mixtures were tested under ambient room temperature 

175 



conditions while Trial 2 mixtures were tested under hot weather conditions of about 

47.8°C(118°F). 

Table 5.10 Mixture Proportioning for Mixture Designs in Phase Ha Testing of 
HVFA and OPC Concrete 



Mixture 


W/C 


Water, 
kg/m 


Cement, 
kg/m 


Fly Ash, 
kg/m 


Coarse 
Aggregate, 

kg/m 


Fine 

Aggregate, 

kg/m 


Air 
Content, 

% 


HWC T1FA 


0.4 


156.6 


195.8 


195.8 


967.0 


728.4 


6 


HWC TIOPC 


0.4 


156.6 


391.6 


0.0 


967.0 


757.8 


6 


HWCT2FA 


0.4 


156.6 


195.8 


195.8 


1008.6 


787.8 


2 


HWC T20PC 


0.4 


156.6 


391.6 


0.0 


1008.6 


787.8 


2 


1 kg/m' =1.68' 


554 lb/yd J 















Coding: HWC = Hot Weather Conditions, T# = Trial number, FA = fly ash, OPC = 
ordinary portland cement 

The temperatures were recorded using the CRIOx datalogger model from Campbell 
Scientific (See Figure 5.9). Type J thermocouples from Omega Engineering Inc. were 
used such that the positive wire was iron and the negative wire was constantan (copper- 
nickel alloy). The insulation around the wires was neoflon (copolymer). The maximum 
temperature that the wires could perform in was 200°C (392°F). The wires were twisted 
together. The twisted ends of the wires were then dipped in liquid tape to help keep the 
wires protected in the concrete while the concrete hardened during curing. It was the 
decision of the author to take temperatures in the middle of the cylinder samples. 
Concrete cylinders were made for the temperature testing. For trials 1 and 2 temperature 
testing, one layer of concrete was placed in the cylinder (about 10.2 cm [4 in] in depth) 
and was consolidated. The wire was then placed about midpoint of this layer and finally 
the last layer of concrete was placed (See Figure 5.10). 



176 




Figure 5.9 Campbell Scientific Datalogger (CR 10X) Used to Record Concrete 
Temperatures 




Figure 5.10 Installing the Thermocouple Into Concrete Sample 

Figure 5.1 1 shows the average internal temperatures recorded for HVFA and OPC 
concrete cured in ambient room temperature conditions. In the Figure the room 
temperature throughout the curing process is provided. The average temperatures are 
also graphed with standard deviations which are shaded in blue for OPC concrete and 
pink for HVFA concrete. From the figure it appears as though HVFA concrete peaks at a 
slower rate compared to OPC and has a peak temperature about 11% smaller than OPC. 



177 




- - FA OPC Ambient 



Figure 5.11 Internal Curing Temperatures of Ambient Cured Fly Ash and OPC 
Samples During Trial 1 Testing 

Figure 5.12 shows the internal curing temperatures when the HVFA and OPC concrete 

trial 2 samples were cured in heated conditions. The curing temperature of about 47.8°C 

(1 18°F) was the maximum temperature that was desired for this study because it 

represented some of the warmest temperatures parts of India have reached in the past 2 

years, although Rajkot has had a high temperature of 42°C (107. 6°F). Under heated 

conditions the HVFA concrete demonstrated that it has a slower rate of heat gain 

compared to OPC concrete. The OPC peaked at about 58°C (136.4°F) which was about 

5% higher than HVFA concrete. Under heated conditions the benefit of lower internal 

temperatures of HVFA concrete might decrease by about 5%. The percentage difference 

between HVFA concrete and OPC did not match those reported by Wang and Yang 

(2006) and Atis (2002) but lower temperatures for HVFA concrete were verified in the 

both tests. Also a unique characteristic of HVFA concrete was observed during the heat 

curing. In Figure 5. 13 the surface of the concrete samples are shown. The surface of the 

178 



HVFA concrete sample develops a glassy sheen compared to the OPC concrete and may 
be attributed to the amorphous phase present in the fly ash. 



60 



50 



40 



U 



2 30 
o 



H 



20 



10 



- 132 



112 









92 >§ 



72 Tl 



52 



32 



10 



Hours 
■ HWC T2 FA HWC T2 OPC 



-Oven 



Figure 5.12 Internal Curing Temperatures of Heat Cured HVFA and OPC Samples 
During Trial 2 Testing 




(a) (b) 

Figure 5.13 Surface of Samples after Heat Curing (a) Fly Ash Mixture (b) OPC 
Mixture 



179 



5.5 Phase III study for HVFA in Hot Weather Conditions: Laboratory Testing of 
Structural and Durability Properties. 

The verification of lower internal temperatures for HVFA concrete compared to 
OPC concrete led to the main phase of the study. Table 5.11 shows the mixture 
proportion for 16 batches made between HVFA concrete and OPC concrete. The mixture 
codes are understood as follows: 

^>^ Coarse Aggregate Content (%) 

OPC55W 

Ordinary Portland Cement (OPC) or s^ ^ ... . .,,, ., ... , 

50% Fly Ash (FA) / ^^Vater (W) or Heat (H) cured 

8 mixtures were not influenced by heated aggregate while the 8 mixtures were. Recall 
that the heated aggregate is meant to represent the possibility of exposing stockpiled 
aggregate to hot weather conditions or the inability to keep even stored aggregate cool in 
extreme temperatures. The water cured samples represented ideal conditions and the heat 
cured samples were cured in temperatures ranging from above 37.8°C (100°F) to 47.8°C 
(1 18°F). Fresh and hardened concrete tests were performed following ASTM standards 
listed in Table 5.12. At least two samples were used for hardened concrete test except 
modulus of elasticity and length change. Modulus of elasticity relied on one new sample 
for each testing day while the length change relied on at least one sample up to 90 days. 

Ideal (water) curing was accomplished with a water tank where the temperature of 
the water was around 22.2°C (72°F). The water tank is shown in Figure 5.14 (a). The 
heated tank was similar to the water tank, customized with bricks (with holes) arranged 
on the bottom of the tank to allow air to circulate fully around the concrete samples. Two 
heaters were placed on either ends of the inside of the tank. The heaters were modified 



180 



space heaters to continuously heat the tank between 37.8°C (100°F) and 47.8°C (118°F). 
Fans were also placed in the tank to allow the air to circulate throughout the curing 
process. No moisture was added to the tank nor to the samples once the samples were 
placed in the tank. The heaters were connected to a timer that allowed the heaters to 
remain on for 6 hours and turn off for 6 hours. 



Table 5.11 Mixture Proportioning for HVFA and 
Extreme Hot Weather Condition Testing 


OPC Concrete Mixture Designs in 


Mixture 


Code 


Heated 

Aggregate 


W/(C+FA) 


Water, 
kg/m 


Cement, 
kg/m 


Fly Ash, 
kg/m: 


Coarse 
Aggregate, 

kg/m 


Fine 

Aggregate, 

kg/rn 


WRA, 
ml/100 kg 


1 


OPC55W 


No 


0.4 


150.7 


376.7 





1014.5 


838.9 


652 


2 


OPC65W 


No 


0.4 


150.7 


376.7 





1195.5 


656.2 


417 


3 


OPC55H 


No 


0.4 


150.7 


376.7 





1014.5 


838.9 


652 


4 


OPC65H 


No 


0.4 


150.7 


376.7 





1195.5 


656.2 


417 


5 


50FA55W 


No 


0.4 


150.7 


188.4 


188.4 


996.7 


800.9 


652 


6 


>0FA65W 


No 


0.4 


150.7 


188.4 


188.4 


1177.1 


622.9 


417 


7 


50FA55H 


No 


0.4 


150.7 


188.4 


188.4 


996.7 


800.9 


652 


8 


50FA65H 1 


No 


0.4 


150.7 


188.4 


188.4 


1177.1 


622.9 


417 


9 


OPC55W 


Yes 


0.4 


150.7 


376.7 





1014.5 


838.9 


1304 


10 


OPC65W 


Yes 


0.4 


150.7 


376.7 





1195.5 


656.2 


417 


11 


OPC55H 


Yes 


0.4 


150.7 


376.7 





1014.5 


838.9 


652 


12 


OPC65H 


Yes 


0.4 


150.7 


376.7 





1195.5 


656.2 


417 


13 


50FA55W 


Yes 


0.4 


150.7 


188.4 


188.4 


996.7 


800.9 


652 


14 


>0FA65W 


Yes 


0.4 


150.7 


188.4 


188.4 


1177.1 


622.9 


417 


15 


50FA55H 


Yes 


0.4 


150.7 


188.4 


188.4 


996.7 


800.9 


652 


16 


50FA65H 


Yes 


0.4 


150.7 


188.4 


188.4 


1177.1 


622.9 


417 


1 kg/m' = 1.68554 lb/yd J 

Table 5.12 ASTM Standards Used for Fresh and Hardened Concrete Tests 


Fresh Concrete Tests Standard Time of Test 


Slump ASTM C 143 Batching 
Unit Weight ASTM C 138 Batching 
Air Content ASTM C 231 Batching 


Hardened Concrete Tests Standard Time of Test 



Compressive Strength 
Rapid Chloride Ion 

Penetrability 

Modulus of Elasticity 

Length Change 



ASTM C 39 

ASTM C 1202 

ASTM C 469 
ASTM C 490 



1,3, 7, 28, 56, 90 days 

28, 56, 90 days 

1,3, 7, 28, 56, 90 days 
1,3, 7, 28, 56, 90 days 



181 



This cyclic heating was meant to represent the thermal gradient that occurs from 
the temperature shift between night and day. The temperature of the tank, when the 
heaters were off, was about 22.2°C (72°F). In fact in Rajkot, India summer nights are 
usually 22.2°C (72°F) or warmer. Figure 5.14 (b) shows how the concrete samples were 
placed in the heated tank. The concrete samples were also covered with 6 mil 
polyethylene sheets to represent field curing when trying to maintain prevent evaporation. 
The heated tank was sealed with aluminum foil insulation and boards to help retain the 
heat in the tank. Figure 5.15 shows the aluminum and boards placed on the tank. Figure 
5.16 is a schematic of the heated tank with dimensions and orientation of the bricks. Two 
tanks were actually customized to simulated heated conditions and two tanks were used 
for the ideal water curing conditions. The heated tanks had to hold a minimum of 152 
samples together and this was similar for the water tanks. 




(a) (b) 

Figure 5.14 View of (a) Water Curing Tank (i.e. Ideally Cured) and (b) Hot 
Weather Curing Tank 



182 



1 






JWI^Bft.MI 






\ 










(a) (b) 

Figure 5.15 Hot Weather Simulation Tank (a) Boards to Keep Heat in (b) Close-Up 
of Aluminum Foil Bubble Insulation 

Top View of Tank 



%m 



■&K 



S« 









%**;£■ 



5*3" 



sat? 



flS ' g ' aa? 



as 



^i-fe 



■*a£ 









is; 



ill 



E.V -J / ^ 






Hliife^ 



• : i?-; 



■fcj! 



?9* 






Mp^H 






:&: 



ac 



SB 

7 <F. i 



15 



m. 



-Bricks 



Space Heaters and Fans 



timtommmim&mMtA 



Side View of Tank 





| 


0.6 


m 






1 ' 




\ 












&m 




m& 


1 

C Cs 


m 




-4.5" 





0.6 m 



0.1 m 



Figure 5.16 Schematic of Hot Weather Simulation Tanks 

In Phase III of the hot weather testing internal temperatures were also recorded 
for the full 90 days of the curing. However, the placement of the wire at midpoint of the 
concrete layer and half the depth of the cylinder was ensured with better precision by 
taping the wire at half the distance of a small dowel that had a length of 20.3 cm (8 in). 
Two cylinders were used to take temperature recordings while the samples cured for 90 
days. In this case the CR5000 Campbell Scientific datalogger was used to keep record of 
temperatures. This datalogger was used because more channels for the thermal couples 



183 



were available. Figure 5.17 shows a picture of the CR5000 datalogger and its placement 
between the heated and water curing tanks. 





(a) 

Figure 5.17 Campbell Scientific (a) Datalogger (CR 5000) and (b) Setup for the Ideal 
and Hot Weather Simulation Tanks for Recording Concrete Temperatures 

Temperatures of the materials were recorded before mixing and reported in Table 

5.13. The concrete temperature was also recorded just after mixing. On average the no 

heated aggregate OPC concrete mixtures were about 24.4°C (76°F) and no heated 

aggregate HVFA mixtures were 22.7°C (73°F). The heated aggregate OPC mixtures 

were about 32.8°C (91°F) while HVFA concrete mixtures with heated aggregate were 

about 30.5°C (87°F). The average temperatures between HVFA and OPC concrete 

provide another demonstration how the HVFA concrete will develop lower temperatures 

over than OPC concrete even with heated aggregate. However, temperature differences 

only range between 6 and 8%. Thus another question arises whether HVFA concrete can 

maintain these percentage differences when actually placed in the field. Peak 

temperatures during the curing process are reported in Table 5.14 for the concrete 

mixtures without the heated aggregate. Peak temperatures for the concrete mixtures with 

heated aggregate have yet to be analyzed. But with the no heated aggregate concrete it is 

verified that HVFA concrete mixtures have lower internal temperatures during hydration. 

184 



However, the difference between HVFA temperatures and OPC temperatures increases to 
approximately 13%. The percentage difference was calculated using the average 
temperature for all OPC and HVFA concrete mixtures. 
Table 5.13 Material Temperatures Before Mixing (And During Mixing for the 



Heated Aggregate Mixtures) 
















Mixture 


Code 


Heated 

Aggregate 


Tee, °F 


Tea, °F 


Tfa, °F 


Tw, °F 


Tfla,°F 


Tconc, 

°F 


Tlab, °F 


Tmixing, 
°F 


1 


OPC55W 


No 


77 


77.9 


77.7 


76.1 


- 


76 


74 


- 


2 


OPC65W 


No 


77 


78 


76.6 


77.3 


- 


75 


70 


- 


3 


OPC55H 


No 


85.1 


78.2 


78.8 


79.1 


- 


82 


76 


- 


4 


OPC65H 


No 


76.3 


76.2 


75.5 


74.1 


- 


71 


70 


- 


5 


>0FA55W 


No 


78.9 


80 


80 


76.2 


79.8 


76 


75 


- 


6 


50FA65W 


No 


77 


75.5 


74 


73.7 


77.3 


72 


75 


- 


7 


50FA55H 


No 


80.4 


80.7 


80.1 


77 


80.2 


76 


74.5 


- 


8 


50FA65H 


No 


73.5 


72 


70 


73.4 


74.3 


69 


70 


- 


9 


OPC55W 


Yes 


75.3 


143 


138.2 


70.8 


- 


88 


73 


108.8 


10 


OPC65W 


Yes 


77.1 


130.2 


140.5 


76.2 


- 


91 


70 


102.5 


11 


OPC55H 


Yes 


74.1 


132.8 


115 


69.2 


- 


90 


65 


98.9 


12 


OPC65H 


Yes 


76.1 


154.7 


143.6 


71.7 


- 


94.5 


68 


105.2 


13 


50FA55W 


Yes 


75.2 


136.2 


136.4 


74.8 


75 


86.5 


68 


95.5 


14 


50FA65W 


Yes 


77 


133.3 


154 


71.9 


76.1 


87 


68 


93.7 


15 


50FA55H 1 


Yes 


68.3 


139.8 


134.2 


70.3 


72.3 


85 


67 


96.6 


16 


50FA65H 


Yes 


76.1 


147.5 


156.3 


69 


76.1 


91 


69 


103.4 



Table 5.14 Internal Peak 1 


remperatures During 


Mixture 


Code 


Heated 
Aggregate 


Temperature 
Measured 

(°C) 


1 


OPC55W 


No 


31.2 


2 


OPC65W 


No 


31.2 


3 


OPC55H 


No 


31.5 


4 


OPC65H 


No 


37.8 


5 


50FA55W 


No 


27.2 


6 


50FA65W 


No 


25.5 


7 


50FA55H 


No 


28 


8 


50FA65H 


No 


33 


9 


OPC55W 


Yes 


TBD 


10 


OPC65W 


Yes 


TBD 


11 


OPC55H 


Yes 


TBD 


12 


OPC65H 


Yes 


TBD 


13 


50FA55W 


Yes 


TBD 


14 


50FA65W 


Yes 


TBD 


15 


50FA55H 


Yes 


TBD 


16 


50FA65H 


Yes 


TBD 



Curing 



185 



5.5.1 Compressive Strength 

Compressive strength tests occurred as early as 1 day of curing. The purpose of 
testing at such an early age was to determine how much of a strength gain concrete 
samples can gain when the hydration is accelerated from hot temperatures. 
Early Strength Gain 

Figure 5.18 shows the early strength gain (average compressive strength results) 
for the first 14 days of curing. In all cases by 14 days OPC strength is much larger than 
HVFA concrete strength. If the largest OPC strength (OPC65H) is compared to the 
smallest HVFA concrete strength (50FA55W) at 14 days for the no heated aggregate 
results OPC has a strength approximately 2 times larger than the HVFA concrete. 
Another observation made in the no heated aggregate results is that the two water cured 
HVFA concrete mixtures do not reach design strength by 14 days but the heat cured 
HVFA concrete mixtures do. The heated aggregate mixtures surprisingly show that the 
lowest HVFA concrete strength (50FA65W) is only about 37% lower than the highest 
OPC concrete strength (OPC65W). However, only one HVFA concrete mixture with 
heated aggregate pass the design strength by 14 days. It was expected that at least the 
water cured HVFA concrete mixtures would not reach design strength since their true 
strength benefits are usually not expected until 56 days of curing. 
Later Strength Gain 

Later strength gain (average compressive strength results) is shown in Figure 
5.19. For the no heated aggregate concrete mixtures OPC remains higher in strength 
compared to the HVFA concrete mixtures. But both the OPC and HVFA heat cured 
samples show either a decrease or leveling off in strength gain. If peak temperatures are 



186 



compared to final temperatures the following observations are made. OPC concrete 
samples decrease in strength by as much as 14% when analyzing OPC55H. HVFA 
concrete samples decrease in strength by as much as 7% when analyzing 50FA65H. The 
heated aggregate concrete samples also show a decrease in strength for the heat cured 
samples. OPC samples decrease by as much as 12% when analyzing OPC65H. HVFA 
samples decrease by as much as 5% when analyzing 50FA65H. 
Compressive Strength Results and Standard Deviations 

Figure 5.20 shows overall average compressive strength results with standard 
deviations. The standard deviations for HVFA concrete do not overlap standard 
deviations for OPC concrete expect perhaps by 90 days of testing. Therefore, it is 
appropriate to indicate that OPC concrete mixtures performed better than HVFA concrete 
mixtures in terms of strength when concrete did not contain heated aggregate. In the 
other case where heated aggregate was involved by 90 days some OPC concrete mixtures 
do have standard deviations that overlap with the HVFA concrete mixtures. But a t-test 
performed for at least the 90 day strength tests reveal that there are no significant results 
indicating that HVFA had higher strengths than OPC concrete mixtures. 

It is important to note that during the compressive strength testing OPC heat cured 
samples to have a rapid break once ultimate strength was reached while heat cured HVFA 
concrete samples tended to break more subtly. The breaking characteristics should be 
studied in more detail because this could be good indicators of how a structure might fail 
or deteriorate over time. Also textures of the concrete begin to differ as curing 
proceeded. Water cured HVFA samples were initially powdery until about 56 days of 
age. But heat cured samples for both HVFA and OPC concrete were not powdery until 



187 



8000 



7000 




Ml 



O 
c 









PC65W 

PC65H 

0FA65W 

B0FA65H 

JPC55W 
DPC55H 



-£ t -- -K 0FA55W 
50FA55H 



-Deisgn Strength 



(a) 



8000 - 



7000 




Ml 



40 



30 



20 



10 



14 



™ r=»=pPC65W 
S? L=fcJoPC65H 



S | '*- I 5 0FA65W 



B0FA65H 



-3 1 * p PC55W 
^L^doPC55H 



50FA55W 
50FA55H 
Design Strength 



(b) 



Figure 5.18 Early Age Compressive Strength (a) No-Heated Aggregate (b) Heated Aggregate 



00 
00 




14 21 28 35 42 49 56 63 70 77 84 
Days 



OPC65W 

OPC65H 

50FA65W 

50FA65H 

OPC55W 

OPC55H 

50FA55W 

50FA55H 

Deisgn Strength 



(a) 



8000 - 



2000 - 



1000 




10 



— ■— 


3PC65W 




nPP65H 






50FA65W 


-*- 


50FA65H 






— *— 


DPC55W 


L^fcbpC55H 


-+- 


50FA55W 





50FA55H 





Design Strength 



14 21 28 35 42 49 56 63 70 77 
Days 



(b) 



Figure 5.19 Later Age Compressive Strength (a) No-Heated Aggregate (b) Heated Aggregate 



00 

^3 



8000 - 




30 og -"-° pc65w 
B- -»-OPC65H 



*- 50FA65W 

X- 50FA65H 
X -*-OPC55W 
^ -»-OPC55H 
■E, -+- 50FA55W 

— 50FA55H 

— Deisgn Strength 



14 21 28 35 42 49 56 63 70 77 84 
Days 



(a) 




OPC65W 
OPC65H 
50FA65W 
50FA65H 
a -»-OPC55W 
B -»-OPC55H 
£, -+-50FA55W 

50FA55H 

Design Strength 



14 21 28 35 42 49 56 63 70 77 84 
Days 



(b) 



Figure 5.20 Compressive Strength Results (a) No-Heated Aggregate, (b) Heated Aggregate 



c 



56 days of age. OPC heat cured samples showed more powdery texture than HVFA 
samples. Early age heat cured samples had more of a rough texture. Appendix D has 
Figures showing the different textures and breaking results for some water and heat cured 
samples. 

5.5.2 Modulus of Elasticity 

The modulus of elasticity was measured because it describes the stiffness of the 
material. The modulus of elasticity is derived from the elastic response of a stress-strain 
curve resulting from compressive strength. By determining the elastic modulus for the 
concrete samples by as early as 1-day the trend in stiffness is seen. Figure 5.21 shows the 
modulus of elasticity results. By 56 days normal weight concrete should can have a 
modulus of elasticity ranging between 14-42 GPa (2000 to 6000 ksi) [Mindess, Young, 
and Darwin, 2003). All mixtures fall within this range of modulus of elasticity. Early 
age modulus of elasticity for the heated aggregate concrete mixtures show much smaller 
stiffness for the HVFA concrete samples compared to the concrete mixtures having no 
heated aggregate. 

5.5.3 Resistance to Rapid Chloride-Ion Penetration 

Rajkot, India is located approximately in the center of the State of Gujarat, and is 
surrounded by the Arabian Sea from the northwest to the southeast corner of the state. 
One of the shortest distances to the sea from Rajkot is about 90 km (56 mi). Although 
Rajkot is not a coastal city, its proximity to the Arabian Sea might lead to the potential 
for salts (sodium chloride) trapped in the surrounding air or in the soil to cause problems 
to any reinforced concrete. The ingress of chloride ions into concrete can lead to the 
breakdown of a passivating iron-oxide film that surrounds any reinforcement that is in the 



191 



4500 
4000 
3500 

63000 

£ 

'1 2500 

en 

3 
o2000 

3 

■g 1500 
1000 
500 





2 


m 


_!_ 


■ 


fi9 


1 


t 


i 


b 


4 






Hi 


A 



10 20 30 40 50 60 70 

Days 



30960 



27520 



24080 



20640 



17200 ~> 



13760 



10320 



3440 



90 100 



♦ OPC55W BOPC65W A50FA55W »50FA65W *OPC55H BOPC65H A50FA55H »50FA65H 



(a) 









BR 


■ 
♦ 


1 


t 


£3000 -1 
| 2500 - 




■ 
• 


t 


i 


t> 


▲ 






A 


1 


• 


§ 


A 




• 





▲ 



30960 

27520 

24080 

20640 

17200 

13760 

10320 

6880 

3440 



10 20 30 40 50 60 70 80 90 100 

Days 



♦ OPC55W BOPC65W A50FA55W »50FA65W OOPC55H BOPC65H A50FA55H »50FA65H 



(b) 



Figure 5.21 Modulus of Elasticity (a) No-Heated Aggregate Concrete (b) Heated Aggregate Concrete 



to 



concrete. The iron-oxide film originally develops with the presence of a high pH or 
alkaline environment created by the concrete. As the pH is reduced from intrusions, such 
as chloride ions, the electrochemical protection to the steel can break down and the 
corrosion process will be activated (Detwiler, Kjellesen, & Gj0rv, 1991). Elevated curing 
temperatures can decrease the strength of concrete and decrease durability properties 
such as resistance to chloride ion penetration. The rapid chloride permeability test 
(RCPT) was performed in this study as an indirect method of determining whether the 
pore structure of the concrete was affected while the samples were cured in elevated 
temperatures and when the temperature of the aggregate was increased. Some authors 
have described the RCPT as a measure of electric conductivity rather than a measure of 
concrete's resistance to chloride ion penetration (Wee, Suryavanshi, Tin, 2000). 
Nevertheless, the RCPT relies on the pore structure of the cement paste matrix and pore 
solution composition (Jain & Neithalath, 2010). The pore matrix develops as more 
hydration occurs. With elevated temperatures the hydration products do not evenly 
distribute causing a coarsening of the pores (Detwiler, Kjellesen, Gj0rv, 1991). Figure 
5.22 shows the permeability apparatus used in this study. 




Figure 5.22 Permeability Testing Setup 

A variety of experimental conditions or compositions used in this study are 
expected to affect the pore matrix of the concrete samples. These conditions include a) 



193 



the use of cement versus fly ash, b) curing in ideal (water/moist) or heated conditions c) 
changing the coarse aggregate content (of total aggregate content) from 55% to 65%, and 
d) using no-heated or heated aggregates. Each of these conditions are expected to affect 
the pore matrix as follows: 

a) The inclusion of Class F fly ash will cause the cementitious properties of the fly 
ash to react slowly in comparison to OPC. Thus the pores of the cementitious 
matrix will fill in more slowly. In addition, a 50% replacement of OPC with FA 
usually requires 56 to 90 days of curing in order to see the benefits of FA which 
can ultimately increase resistance to chloride ion penetration. 

b) Water cured specimens should allow for a more uniform distribution of hydration 
products in comparison to heat curing. Additionally, heat curing could cause 
water within the samples to evaporate too quickly, thus leaving larger pores and 
eventually leading to more microcracking than what would be seen in water cured 
samples. 

c) The total aggregate content will not change within the design of the mixture, 
however the total coarse aggregate content will increase from 55% to 65%. It can 
be expected that with more coarse aggregate there will be more heat storage 
especially for those samples cured in heat. With more heat the samples should 
again experience a non-uniform distribution of hydration products and the 
possibility of weaker interfacial transition zones (ITZ) surrounding the coarse 
aggregate. 

d) If aggregates are heated, the size and amount of pores within the concrete matrix 
could increase at a much earlier age than the samples without heated aggregates. 



194 



Finally the results from the RCPT test should show that heating the aggregates 

and curing them in heated conditions could show an increase in permeability by 

twice that of the samples without heated aggregates. 

Figures 5.23 (a) through 5.23 (b) both show the average results of the RCPT on 
the concrete samples that were subjected to ideal (moist) and heated curing conditions. 
Additionally, Figure 5.23 (a) represents the samples without heated aggregate while 
Figure 5.23 (b) shows the heated aggregate samples. 
Effect of Curing 

In Figures 5.23 (a) and 5.23 (b) moisture cured specimens showed a decrease in the 
chloride penetration for both the OPC and FA samples. At 28-days of age, both the OPC 
and FA moist cured samples initiated with moderate chloride penetration (2000 to 4000 
coulombs) and by 90 days of age, both mixture types had decreased to low chloride 
penetration (1000 to 2000 coulombs). 

The heat cured specimens did not always show a decreasing trend in chloride 
penetration. Typically, the OPC and FA samples under heated conditions had higher 
permeability readings from the 28 to 90 days of age in comparison to the water cured 
samples. By the 90 days of age, the coulombs ranged between 2870-6517 for the heat 
cured samples, thus the heated samples were classified as having moderate to high 
permeability. 
Effect of Cementitious Material 

When comparing the OPC to the FA samples in Figure 5.23 (a) and Figure 5.23 
(b), the FA samples cured in water initially had higher permeability than the OPC 
samples. In Figure 5.23 (a) 28 day permeability for the water cured FA samples began to 



195 



approach the permeability of the OPC samples. By 90 days of age, all water cured FA 
samples showed a lower permeability than the OPC water cured samples and fell within 
the classification of low permeability (See Figures 5.23 (a) and 5.23 (b)). 

Concrete containing fly ash subjected to heat curing generally had lower 
permeability than the heat cured OPC samples at 28 day testing and remained lower than 
the OPC samples until the 90 day testing. However, all heat cured samples, whether the 
samples contained fly ash or not, exhibited a permeability that was either moderate or 
high. 
Effect of Aggregate Content 

Referring to Figures 5.23 (a) and 5.23 (b) for samples containing 55% or 65% 
coarse aggregate of total aggregate, there was negligible difference in permeability. The 
samples first measured moderate permeability at 28 days of age and eventually all 
resulted in low permeability. 

The heat cured samples showed a larger difference in performance among 55% 
and 65% coarse aggregate. However, it was surprising that the permeability for the 65% 
coarse aggregate samples was lower than the 55% coarse aggregate samples when cured 
in heat. As discussed earlier it was expected with more aggregate there would be more 
heat storage thus affecting the distribution of the hydration products throughout the 
concrete sample and increasing the pores in the cementitious matrix. 
Effect of Heated Aggregate 

Overall there was little difference between concrete mixtures containing the no- 
heated and heated aggregate. An outlier from Figure 5.23 (b) shows the OPC55H mixture 
with heated aggregate having at least a 50% higher permeability than its companion 



196 




OPC55W 50FA5SW OPC5SH 50FA55H OPC65W 50FA65W OPC65H 50FA65H 

□ 28-day ■ 56-day □ 90-day 

(a) 



High 



Moderate 

Low 
Very Low 



o " 

"3 

o 

■o 



■o" 

< 




I 



31L8 



1 
-1808 




OPCS5W S0FA55W OPC55H S0FA55H OPC65W 50FA65W OPC6SH 50FA6SH 



□ 28-day "56-day □ 90-day 

(b) 



Figure 5.23 Average Rapid Chloride Ion Permeability Test Results (a) No-Heated Aggregate, (b) Heated Aggregate 



-J 



mixture, OPC55H, from Figure 5.23 (a) without heated aggregate. In addition, there was 
the case where a heated aggregate mixture resulted in a lower permeability compared to 
the no-heated aggregate mixture by 90 days of age (i.e. OPC55W). The majority of 
heated aggregate mixtures were approximately 10% higher than the companion mixtures 
with no-heated aggregate. 
5.5.4 Length Change 

Figure 5.24 shows the apparatus used to measure length change. Figures 5.25 
through 5.26 show the percentage length change for the no-heated and heated aggregate 
samples respectively. Evening and morning changes were recorded due to the cyclic 
heating conditions created to represent diurnal temperatures for the day. Maximum 
temperatures were approximately 48°C (118°F) and minimum temperatures were about 
22°C (72°F) to represent night temperatures. The lengths of the samples were measured 
in the morning before the heater turned on and in the evening after hours of exposure to 
maximum temperatures. 




Figure 5.24 Length Change Apparatus 

Comparing similar mixtures that experience different curing conditions shows 
that all heat cured concrete samples experienced more length change than the water cured 
samples. Most heat cured samples decreased in length. However, in Figure 5.25 (d) the 

198 



50FA65H mixture did increase in length before 28 days and suddenly at 56 days. 
Maximum percentage length change for heat cured samples was about 0.06% for OPC 
samples and about 0.02% for FA samples in Figure 5.25. In Figure 5.26 maximum 
percentage length change for heat cured samples was about 0.08% for OPC samples and 
0.04% for FA samples 

A comparison of the percentage length change in samples with different 
cementitious materials reveals that FA samples were usually 0.04% less than the OPC 
samples when heat cured. When water cured, the FA samples seem to have a slightly 
more linear expansion than the OPC samples (i.e. 50FA65W versus OPC65W in Figures 

5.25 (d) and 5.25 (c) respectively and 50FA55W versus OPC55W in Figures 5.26 (b) and 

5.26 (a) respectively). 

Aggregate content did not seem to affect FA samples as much as OPC samples. 
In Figure 5.25 the OPC samples differed by 0.03% while FA samples differed by 0.018% 
when comparing aggregate content. In Figure 5.26 OPC samples differed by 0.006% and 
FA samples differed by 0.002%. However, these differences did not necessarily always 
mean the 65% coarse aggregate mixtures had higher length change than the 55% coarse 
aggregate mixtures. 

The heated aggregate samples did have higher percentage length changes than the 
no-heated samples. Maximum differences that occurred among the OPC samples were 
0.04% and 0.028% for the FA samples as a result of heat curing. 
Overall FA samples appeared to have less of a change in length compared to OPC 
samples. These changes in lengths in terms of units of length were not large. Maximum 
length change was about 0.02 cm (0.008 in). 



199 



s 

u 



0.10 
0.08 
0.06 
0.04 



g 0.02 
g 0.00 
£ -0.02 

OJO 

g -0.04 

J -0.06 

-0.08 

-0.10 



(iTL-w-^uir 



-f * 



-60- 



-80- 



OPC55W evening 
OPC55W morning 
OPC55H evening 
OPC55H morning 



Days 



(a) 




■OPC65W evening 
-OPC65W morning 
■OPC65H evening 
■OPC65H morning 



Days 



(b) 



o 
an 
c 
a 

J3 

u 

J3 
*^ 

m 



0.10 
0.08 
0.06 
0.04 
0.02 
0.00 
-0.02 
-0.04 
-0.06 
-0.08 
-0.10 



(iW^"Sr^^~-Ma- 



=f ■*■ 



50FA55W evening 
50FA55W morning 
50FA55H evening 
50FA55H morning 



Days 



(C) 




-50FA65W evening 
■50FA65W morning 
-50FA65H evening 
-50FA65H morning 



Days 



(d) 



Figure 5.25 Length Change for No-Heated Aggregate Samples 



to 

o 

o 




Days 



■OPC55W evening 
-OPC55W morning 
■OPC55H evening 
■OPC55H morning 



(a) 




Days 



■OPC65W evening 
■OPC65W morning 
■OPC65H evening 
■OPC65H morning 



(b) 




Days 



-50FA55W evening 
■50FA55W morning 
-50FA55H evening 
■50FA55H morning 



(C) 




Days 



(d) 



- 50FA65 W evening 
■ 50FA65W morning 
-50FA65H evening 
■50FA65H morning 



Figure 5.26 Length Change for Heated Aggregate Samples 



to 

o 



5.6 Applying a Multiple Linear Regression Model to Determine the Significance of 
Testing Variables on HVFA Concrete versus OPC Concrete When Subjected to Hot 
Weather Conditions 
5.6.1 Background on Multiple Linear Regression 

Multiple linear regression models illustrate the relation between the dependent (response) 
variable and the independent (predictor) variables based on a regression equation (Hayter, 
1996). The general form of the multiple regression equation with k variables is the 
following: 

yi=fiO + fil x il + fi2 x i2 + - + Pk x ik+ e i> i = l,2,...,n (1) 

fto is considered the intercept parameter and /?, is the parameter that determines how the 
input variable, x it has an influence on the response variable while all other input variables 
are fixed. /?o ,. . ., ^is estimated using the method of least squares and are chosen so that 
the sum of the squares of the vertical distance between the actual observation and the 
fitted values is minimized. The null hypothesis is Hq: fii = ... = /?* =0. The alternative 
hypothesis is H A : /?, ^0. If fii=0 then the input variable xt has no influence on the response 
variable and can be left out of the model. If the null hypothesis is rejected then x t has 
some influence on the response variable and should be included in the model. The 
hypotheses are compared to a ^-distribution such that the degrees of freedom are 
calculated from n-k-l(n = samples size, k+l= the number of parameters). A two sided p- 
value (measure of plausibility) is calculated such that 
p-value = 2 x P(X > t) 

and X is a random variable with a ^-distribution with n-k-1 degrees of freedom. A list of 
p-values will be obtained that correspond to the parameters /? , . . ., /?*> The p-values are 



202 



important for all parameters except fio- The input variables x\ corresponding to the 
parameter/?,, usually, should not be included in the model if the p-value is larger than 
10%. If the p-value is smaller than 1% than the input variable is considered important to 
the model. If the p-value is between 1% and 10% it is not obvious whether the input 
variable is important to the model and the decision to keep it is left up to the 
experimenter's judgment (Hayter, 1996). 
5.6.2 Application of the Multiple Linear Regression Models 
Compressive strength, permeability, and percentage length change were measured at 
common time intervals (28, 56, and 90 days). Initially each of the following conditions 
were expected to linearly affect the three measurements: (1) curing conditions, (2) 
aggregate content, (3) time of curing, and (4) temperature of aggregate. Therefore 
strength, permeability or percentage length change is expected to be the result of the sum 
of the various conditions (each applying a certain level of influence). However, the 
degree of influence was unknown. 

Therefore the goal of a multiple regression analysis for the study performed on 
the OPC and high volume fly ash concrete samples was to the determine the effects on 
the measured dependent variables (X, Y and Z) as a result of a variety of experimental 
conditions and compositions or independent variables (A, B, C, D, and 7) such that 



X = compressive strength 

Y = permeability 

Z = length change as a percentage 



>- Dependent 
Variables 



203 



Independent 
Variables 



A = cementitious material (cement vs. fly ash) \ 

B = curing condition (water vs. heat) 

C = aggregate content (55% vs. 65% 

coarse aggregate content of total aggregate) 

D= heated or not heated aggregate 

T =three common points in time 

(28, 56, and 90 days of curing) J 

The three multiple linear regression models were designated as 

X = P Q + ^A+ fcB+ frC+ faD+ frT (2) 

Y=fi + J3jA+ p 2 B+ p 3 C+ p 4 D+ fcT (3) 

Z = P Q + ^A+ fcB+ frC+ faD+ fcT (4) 

Table 5.15 is a matrix showing the values gathered from testing compressive strength, 
permeability, and length change. The independent variables are either represented as a 1 
or to indicate that two cases (e.g. cement or fly ash, heat or non heated aggregate) were 
tested within each independent variable. The independent variable T, however, is 
identified as the actual number of days of curing (i.e. 28, 56, and 90). The independent 
variable (T*B) will be explained in a later section. 

Equations (2) through (4) were evaluated using the statistical package Minitab. 
Minitab results are shown in Appendix D. Figure D.18 in the Appendix provides an 
explanation of the different parameters calculated within the regression analysis (these 
explanations are the bolded items in Figure D.18). The equations of the fitted curves for 
compressive strength, permeability and percent length change are shown in Table 5.16. 
Table 5.17 provides a summary of the results from Appendix D. The equations or 



204 



response functions represent a hyperplane (a plane in more than three dimensions). 
Although it is difficult to picture the response functions, the meaning of the parameters 
(coefficients) can be understood as follows: Referring to Equation (1), a unit increase in 
the independent variable Xk, with all other independent variables held constant, means 
that the mean response E {yj will change based on the parameter/^. 
Table 5.15 Matrix for Multiple Linear Regression Analysis 





Dependent Variables 






Independent Variables 








Case ID 


X 


Y 


Z 


A 


B 


C 


D 


T 


T*B 




psi 


coloumbs 


% (0-cement, 1 


-fly ash) (0-water, ] 


-heat) (0-55%, 1-65%) (0-not-heated, 1-heated) 


Days 




OPC55W 


6369.67 


2756.69 


0.0030 














28 





OPC55W 


7162.67 


2417.35 


0.0020 














56 





OPC55W 


7369.67 


1879.01 


0.0010 














90 





OPC55H 


7005.00 


4304.02 


-0.0210 





1 








28 


28 


OPC55H 


7007.00 


3661.44 


-0.0270 





1 








56 


56 


OPC55H 


6241.33 


4276.95 


-0.0358 





1 








90 


90 


OPC65W 


6629.67 


3322.10 


0.0010 













28 





OPC65W 


6734.67 


2547.76 


0.0010 













56 





OPC65W 


7209.00 


1800.49 


0.0045 













90 





OPC65H 


7566.67 


3749.89 


-0.0415 





1 







28 


28 


1 OPC65H 


7652.67 


3451.16 


-0.0495 





1 







56 


56 


| OPC65H 


7604.00 


3837.43 


-0.0590 





1 







90 


90 


i 50FA55W 


4488.33 


3382.57 


0.0030 













28 





Z 50FA55W 


4975.67 


2485.94 


0.0020 













56 





50FA55W 


5518.33 


1360.52 


0.0020 













90 





50FA55H 


5220.00 


3844.65 


-0.0153 




1 








28 


28 


50FA55H 


4634.33 


3677.69 


-0.0228 




1 








56 


56 


50FA55H 


4800.33 


3415.06 


-0.0252 




1 








90 


90 


50FA65W 


4582.33 


3616.77 


0.0090 












28 





50FA65W 


5229.33 


2493.16 


0.0075 












56 





50FA65W 


5602.67 


1524.32 


0.0110 












90 





50FA65H 


5798.00 


2415.54 


0.0185 




1 







28 


28 


50FA65H 


5301.67 


3170.93 


0.0088 




1 







56 


56 


50FA65H 


5569.33 


2870.40 


-0.0088 




1 







90 


90 


OPC55W 


6935.50 


2728.71 


0.0040 













28 





OPC55W 


7107.50 


2078.46 


0.0060 













56 





OPC55W 


7651.00 


1835.69 


0.0065 













90 





OPC55H 


5100.00 


2981.86 


-0.0557 





1 







28 


28 


OPC55H 


5774.00 


6068.86 


-0.0615 





1 







56 


56 


OPC55H 


5654.50 


6516.50 


-0.0680 





1 







90 


90 


OPC65W 


7103.00 


2327.10 


0.0030 












28 





OPC65W 


8062.00 


1846.97 


0.0045 












56 





OPC65W 


7985.00 


1869.98 


0.0060 












90 





OPC65H 


6133.00 


4062.60 


-0.0427 





1 






28 


28 


OPC65H 


7193.00 


4329.74 


-0.0498 





1 






56 


56 


| OPC65H 


6462.50 


4250.32 


-0.0577 





1 






90 


90 


X 50FA55W 


4733.50 


2979.15 


0.0170 












28 





50FA55W 


5452.50 


2415.09 


0.0205 












56 





50FA55W 


6039.00 


1497.70 


0.0225 












90 





50FA55H 


4691.00 


3117.69 


-0.0228 




1 







28 


28 


50FA55H 


4545.00 


4239.95 


-0.0242 




1 







56 


56 


50FA55H 


4296.00 


3891.58 


-0.0280 




1 







90 


90 


50FA65W 


4724.00 


3482.75 


0.0040 











28 





50FA65W 


5342.00 


2442.62 


0.0055 











56 





50FA65W 


5925.50 


1808.16 


0.0055 











90 





50FA65H 


5344.50 


2309.05 


-0.0230 




1 






28 


28 


50FA65H 


5194.50 


2176.83 


-0.0283 




1 






56 


56 


50FA65H 


5053.50 


3070.31 


-0.0305 




1 






90 


90 



205 



Table 5.16 Equations of Fitted Curves from 1 st Regression Analysis 

Regression Analysis X , Y, and Z versus A, B, C, D, T 
Xi = 6563-1777 A-379 B+468 C-157 D+6.48 T 
Yj = 3084-467 A+1366 B-377 C+86 D-5.77 T 
Zi = 0.00569+0.0181 A-0.0384 B+0.00070 C-0.00649 D-0.000095 T 

Table 5.17 Summary of 1 st Regression Analysis 



Dependent 
Variable 


Significant 
Variables 


Insignificant 
Variables 


R 2 


Standard Deviation 


Xi 


A,B,C 


D, T 


78.2% 


540.204 psi (3.72 MPa) 


Yi 


B 


A, C, D, T 


49.9% 


815.909 coulombs 


Zi 


A, B 


C,D,T 


77.3% 


0.0125 % 



For compressive strength all p-values were less than 10% except for the 
independent variable D. This indicates that the variable D (aggregate was heated or not 
heated) is not needed in the model to improve the fit. All other p-values that are below 
10% indicate that the null hypothesis is not plausible and thus the strength has a 
relationship with the type of cementitious material (A), curing environment (5), coarse 
aggregate content (C), and number of curing days (7). 

The outputs for permeability (Refer to Figure D.19) show that the variables C, D, 
and T are not significant within the model. This confirms the observational results 
discussed in the permeability section However, as noted from Figure 5.22a and 5.22b 
OPC55H (heated and no-heated aggregate) samples were outliers, having differences in 
permeability up to 50% at 90 days of curing. 

The regression analysis for percentage length (Refer to Figure D.20) change 
revealed that coarse aggregate content (C) and curing time (7) are not of statistical 
significance within the model. Again this confirms the observations made in the length 
change section. However, from Figures 5.24 and 5.25 it can be seen that over time the 
percentage length change usually results as shrinkage for the samples cured in heated 



206 



conditions, while the samples cured in water did not have such large changes in length 
over time. These differences may be a reason why the regression analysis showed that 
curing time (7) had no major effect on the samples length change. 

Within linear regression it is customary, upon reviewing the results of the initial 
regression analysis, to remove the variables that were not significant to each model and to 
perform the analysis again. However, the first analysis was completed assuming that the 
predictors or the independent variables were additive. Therefore, it was within the 
interest of the author to run the same analysis with the inclusion of an interaction between 
the independent variables T and B. This interaction term is also known as a bilinear term. 
It was assumed that the interaction would be important in developing a better model to fit 
the data for compressive strength, permeability, and percentage length change. The 
interaction was believed to exist because the level of change in the dependent 
variable is determined by the adjoining effect of T changing and B changing in the 
model. In other words, using strength as an example, it is expected that a sample should 
gain more strength with time but that strength should decrease by some factor if the 
sample was cured in heated conditions or the strength should increase if the sample was 
cured in water. 

The previous three multiple linear regression equations (2 through 4) changed to 
include the interaction term as follows. 

X = B + BjA+ p 2 B+ p 3 C+ M>+ PsT+fcTB (5) 

Y = P +P ] A+ fcB+ fcC+ p 4 D+ p 5 T + B 6 TB (6) 

Z = P Q + /J ; A+ fcB+ frC+ p 4 D+ p 5 T+ fcTB (7) 



207 



Equations (5) through (7) were evaluated using the statistical package Minitab. Minitab 
results are shown in Appendix D as Figures D.21 through D.23. Table 5.18 shows the 
regression equations resulting from the second analysis and Table 5.19 provides a 
summary of the results from Appendix D. A comparison of Table 5.18 and 5.19 shows 
how the interaction term changes the coefficients /?#, /?2, and /?5. 

Table 5.18 Equations of Fitted Curves from 2 nd Regression Analysis 

Regression Analysis X , Y, and Z versus A, B, C, D, T, T*B 

X 2 = 6042-1777 A-663 B+468 C-157 D+15.5 T-18.0T*B 

Y 2 = 4032-467 A-529 B-377 C+86 D-22.1 T+32.7T*B 

Z 2 =-0.00160+0.0181 A-0.0239 B+0.00070 C-0.00649 D+0.000030 T-0.000251 T*B 



Table 5.19 Summary of Regression Analysis When Including the TB Interaction 
Term 


Dependent 
Variable 


Significant 
Variables 


Insignificant 
Variables 


R 2 


Standard Deviation 


X 


A, B, C, T, TB 


D 


82.6% 


488.115 psi (3.36 MPa) 


Y 


A, B, T, TB 


C,D 


64.6% 


693.672 coulombs 


Z 


A, B 


C, D, T, TB 


79% 


0.0122 % 



The effects of the coefficients and the fit that they provide to the model can be 
examined by looking at just the (fisT+ faTB) term from equations (5) through (7). 
Assuming all other independent variables are constant and comparing the response 
variable (dependent variable) with the predictor (independent variable) T the following 
should be expected with the (fisT+ faTB) term: 

Compressive strength X 

The term fi 5 T+ fi 6 TB is ( 1 5.5 -18.0B)*T from X 2 in Table 5.17 

IfB = (water cured), then the slope of the term is 15.5, which means strength 
should increase as time increases. 

IfB= 1 (heat cured), then the slope of the term is -2.5, which means strength 
should decrease as time increases. 



208 



To provide confirmation of the two different slopes the average strengths 
resulting from the regression equation X2 are graphed against the average of the 
real measured strengths from Table 5.15, when B=0 and B=l at T = 28, 56, and 
90 days. This graph is shown in Figure 5.27. 






BO 

=5 

QJ 
'— 

in 

QJ 

"S 

09 

•- 



o 
U 



7500 
7000 
6500 
6000 
5500 
5000 
4500 
4000 













< 














- 


1 1 1 



52 
46 
40 
34 



o 

o 



13 

re 

re 

3 

S3 

era 



T3 

28^ 







28 



56 



84 



Days 



■Water Cured 
Estimated Heat Cured 



•Heat Cured 
'Estimated Water Cured 



Figure 5.27 Effects of the Interaction of T and B on Compressive Strength 

Permeability Y 

The term p 5 T+ p 6 TB is (-22.1 +32.7B)*T from Y 2 in Table 5.17 

IfB = (water cured), then the slope of the term is -22.1, which means 
permeability should decrease as time increases. 

IfB= 1 (heat cured), then the slope of the term is 10.6, which means permeability 
should decrease as time increases. 

To provide confirmation of the two different slopes the average permeability 
values resulting from the regression equation Y2 are graphed against the average 
of the real measured permeability values from Table 5.15, when B=0 and B—l at 
T = 28, 56, and 90 days. This graph is shown in Figure 5.28. 



209 



s 

o 

S 
O 

U 



cS 

I 

•— 

Oh 



5000 
4500 
4000 
3500 
3000 
2500 
2000 
1500 
1000 
500 





28 



56 



84 



Days 



•Water Cured ^- Heat Cured 

'Estimated Water Cured ^*- Estimated Heat Cured 



Figure 5.28 Effects of the Interaction of T and B on Permeability 

Percentage Length Change Z 

The term fi 5 T+ fi 6 TB is (0.000030-0.00025 1B)*T from Y 2 in Table 5.17 

IfB = (water cured), then the slope of the term is 0.000030, which means 
percentage length change should increase as time increases. 

IfB= 1 (heat cured), then the slope of the term is -0.000221, which means 
permeability should decrease as time increases. 

To provide confirmation of the two different slopes the average percentage length 
change resulting from the regression equation Z 2 are graphed against the average 
of the real measured percentage length change values from Table 5.15, when 
B=0 and B=l atT = 28, 56, and 90 days. This graph is shown in Figure 5.29. 

The interaction of A and B was also considered but the R values for the models 

did not increase as much as they did with the interaction between T and B, except for the 

model for permeability (Z). Nevertheless, the decision was made that the T and B 

interaction was the only interaction that would be used to develop regression models. 

The influence of A (descriptor cement or fly ash) on all regression analyses, however, 



210 



was significant in all models. Thus this provides evidence that compressive strength, 
permeability, and percentage length change will vary with use of cement or fly ash. 



A AOAA 






U.UjUU 




0.0200 - 












o 0.0100 
^ 0.0000 








' 


" 








DJD . 

£ -o.oioo ( 

W -0.0200 


) 28 56 «.! 














2F -U.U3UU 








j -0.0400 - 










-U.UjUU 






-U.UOUU 

Days 




^^ Water Cured ^^ Heat Cured 




^^Estimated Water Cured ^*- Estimated Heat Cured 





Figure 5.29 Effects of the Interaction of T and B on Percent Length Change 
5.6.3 Revision of Multiple Linear Regression Analysis with Original Data 

Previously, the multiple linear regression model was developed with the average 
values. For example, for each testing day the compressive strength (28, 56, and 90 day 
testing) was the result of at least two samples averaged and a linear regression equation 
was fitted to these average values. However, using solely the averages of the data did not 
take into account the variability within each testing day. Therefore the multiple linear 
regression analysis was carried through with the inclusion of all original data. Below is a 
discussion of the resulting coefficient of determination (R 2 ) and standard deviation of 
each of the multiple linear regression equations. 



211 



Compressive Strength 

The first regression analysis is not included in this comparison because it did not 

include the interaction terms T*B. X2 represents the regression analysis with the 

averages and X3 is the regression equation which resulted from all original data. The 

influence of each of the input variables (A through TB) changed. This is apparent when 

comparing the coefficients in each of the equations in Table 5.20. 

Table 5.20 A Comparison of Equations of Fitted Curves From 2 nd and 3rd 
Regression Analysis for Compressive Strength 

Regression Analysis X , Y, and Z versus A, B, C, D, T, T*B 
X 2 = 6042-1777 A+663 B+468 C-157 D+15.5 T-18.0T*B 
X 3 = 6014-1802 A+804B+452C-157D+15.3T-18.4T*B 

In Table 5.21 the variables that are significant are shown. The variables were 

considered significant if the p-values were less than 0.05. Both analyses resulted in the 

same variables being significant. The coefficient of determination (R ), however, 

decreased. So with the inclusion of original values the analysis revealed that the spread 

of the compressive strength data made it a little more complex to fit a curve to the data. 

However, with more variability in the data, this resulted in a larger standard deviation for 

estimating a linear regression equation. With the new standard deviation, at least a little 

more than 2/3 of the original data (approximately 69%) falls within one standard 

deviation. 

Table 5.21 Comparing Significant Variables, R 2 , and Standard Deviations for 
Compressive Strength 



Dependent 
Variable 


Significant 
Variables 


Insignificant 
Variables 


R 2 


Standard Deviation 


x 2 


A, B, C, T, TB 


D 


82.6% 


488.115 psi (3.36 MPa) 


x 3 


A, B, C, T, TB 


D 


76.9% 


554.557 psi (3.82 MPa) 



212 



Note regarding the standard deviation in the multiple linear regression analysis: The 

linear regression analysis calculates a standard deviation for all compressive strength 

values including all results for HVFA and OPC together. As was seen in the compressive 

strength results in Figure 5.19, the differences in strength between OPC and HVFA are 

taken into account with the standard deviation calculated by the multiple linear 

regression. In other words the standard deviation is representing the spread of the 

strength results between HVFA and OPC. This is also true for all standard deviations 

reported for permeability and length change. 

Permeability 

In Table 5.22 the multiple linear regression equations for permeability did not 

change despite including twice the set of data that was included in the first analysis. 

However, in Table 5.23 it is clear that R decreases and standard deviation increases. 

With the new standard deviation approximately 76% of the data falls within one standard 

deviation of the multiple linear regression analysis. 

Table 5.22 A Comparison of Equations of Fitted Curves From 2 nd and 3rd 
Regression Analysis for Permeability 

Regression Analysis X , Y, and Z versus A, B, C, D, T, T*B 
Y 2 = 4032-467 A-529 B-377 C+86 D-22.1 T+32.7T*B 
Y 3 = 4032 -467A-529 B-377 C+86 D-22.1 T+32.7T*B 

Table 5.23 Comparing Significant Variables, R 2 , and Standard Deviations for 
Permeability 



Dependent 
Variable 


Significant 
Variables 


Insignificant 
Variables 


R 2 


Standard Deviation 


Y 2 


A, C, T, TB 


B, D 


64.6% 


693.672 coulombs 


Y 3 


A, C, T, TB 


B, D 


54.4% 


823.454 coulombs 



Length Change 

In Table 5.24 the multiple linear regression equations for length change show how 
the fitted equations did not change because the amount of data for length change did not 

213 



increase. The number of samples per mixture remained at 1 per mixture. Thus in Table 

5.25 the R and standard deviations remained the same as well. This meant that 73% of 

the original data fell within one standard deviation of the estimated data. 

Table 5.24 A Comparison of Equations of Fitted Curves from 2 nd and 3rd 
Regression Analysis for Length Change 

Regression Analysis X , Y, and Z versus A, B, C, D, T, T*B 
Z 2 = -0.00160+0.0181 A-0.0239 B+0.00070 C-0.00649 D+0.000030 T-0.000251 T*B 
Z 3 = -0.00160+0.0181 A-0.0239 B+0.00070 C-0.00649 D+0.000030 T-0.000251 T*B 

Table 5.25 Comparing Significant Variables, R 2 , and Standard Deviations for 
Permeability 



Dependent 
Variable 


Significant 
Variables 


Insignificant 
Variables 


R 2 


Standard Deviation 


z 2 


A, B 


C, D, T, TB 


79% 


0.0122 % 


z 3 


A, B 


C, D, T, TB 


79% 


0.0122 % 



Checking the Validity of the Model 

The question to consider is the following: 

Is there at least one independent variable linearly related to the dependent variable ? 

To answer this question the following hypothesis can be tested: 

H :Pi=p 2 = ...=Pk = 

Hi: At least one pi is not equal to zero 
If at least one [3; is not equal to zero, the model has some validity. The hypothesis is 
tested using the method of Analysis of Variance (ANOVA). The analysis of variance 
was completed for the results of the multiple linear regression analysis using Minitab. A 
F-statistic was calculated for each linear regression analysis by dividing the mean squares 
by the mean squares error. The p-value was also determined. As such either the F- 
statistic or the p-value can be used to reject or accept the null hypothesis Ho. The F- 
statistic and p-value are shown in Table 5.26 for the linear regression analysis for 



214 



strength, permeability and length change. If the p-values shown in Table 5.26 are 
compared to an a = 0.05 (representing a 95% confidence interval) it is obvious that the p- 
values are much smaller than 0.05. Thus, this also represents a large F-statistic and if F 
is large the multiple linear regression model is considered valued for each response 
variable (strength, permeability, and length change). The null hypothesis is rejected and 
at least one of the P; is not equal to zero and at least one independent variable 
(cementitious material, curing condition, heated aggregate, aggregate content, days of 
curing) is linearly related to compressive strength, permeability, and length change. 
Table 5.26 Summary of ^-Statistic and P- Value from ANOVA 



Compressive Strength 



F= 62.17 



p = 0.000 



Permeability 



F= 17.73 



p = 0.000 



Length Change 



F = 25.64 



p = 0.000 



Summary 

Overall, the results of including all original data allowed for the coefficients in the 
multiple linear regressions to not change by much or not change at all. However, small 
changes in the R and standard deviation showed how the fitted curves did not allow for 
all the original data to be accurately represented. Nevertheless, at least 2/3 of the 
compressive strength, permeability, and length change data remained in at least one 
standard deviation of the estimated data. Finally the multiple linear regression was model 
was validated using ANOVA and revealed that at least one of the independent variables 
is linearly related to the response variable. 

5.7 Summary of Strength, Permeability, Length Change, and Multiple Linear 
Regression. 

Although high temperatures have been addressed for making, placing, and curing 
concrete the subject of a high temperature environments along with effects of heated 

215 



aggregate and amount of aggregate is not well addressed in research. This dissertation 
presented evidence that fly ash can be beneficial to concrete for maintaining lower 
internal concrete temperatures, however when you have combined effects such as heated 
aggregate, and varying aggregate content and curing in hot weather conditions then 
strength results revealed that OPC will still perform better than HVFA concrete mixtures 
(based on the particular mixture design used in this dissertation). Nevertheless, HVFA 
and OPC concrete mixtures remained above design strength for the 90 days of curing. For 
permeability OPC and HVFA mixtures showed decreases in permeability if water cured. 
If heat cured the permeability reached moderate permeability. For heat cured OPC 
mixtures permeability fell in to the high permeability range. The percentage length 
change in water cured samples was very close to zero. HVFA heat cured samples did 
exhibit some shrinkage but about 0.02% less than OPC heat cured samples. Overall 
HVFA concrete samples performed better or similarly to OPC concrete in terms of length 
change. 

From the multiple linear regression analyses the goal was to determine if 
cementitious material, aggregate content, aggregate temperature, heat curing, and curing 
days (these are considered independent variables) were statistically significant towards 
the results of strength, permeability, and length change. P- values were calculated for 
each independent variable and they were compared to an a = 0.05. However, not all 
independent variables proved to be significant for the dependent variables measured. 
Strength, for example, was significantly affected from the variables cementitious material 
used, curing conditions used, aggregate content, and number of curing days. Heated or 
no heated aggregate did not have much of an effect on either of the mixtures for strength, 



216 



permeability, and length change. It can be concluded that HVFA should be considered 
for extreme hot weather conditions but results also revealed that OPC concrete can be just 
as beneficial in these conditions. 



217 



6. Conclusions and Recommendations 
6.1 Conclusions 

With the production and use of concrete green house gases (GHG) are released 
into the atmosphere. GHGs have been connected with the onset of climate change 
events. The events could lead to varying rain events that lead to flooding, and extreme 
high temperatures that can affect the performance of concrete infrastructure today. 
However, with concrete still having an important role in today's infrastructure there is a 
need to research if attributes of concrete technologies can contribute to carbon mitigation 
and climate adaptation. The purpose of this dissertation was to evaluate Pervious and 
HVFA concrete's structural and environmental properties that could contribute to carbon 
mitigation and climate adaptation in cities with Rajkot as a case study. 
6.1.1 Carbon Mitigation: An MFA-LCA Approach 

The CMA reports a cement emission factor (0.83 tonnes CCVtonne cement) very 
close to the calculations performed in this disseration for the state of Gujarat (0.84 tonnes 
CCVtonne cement). However, it was necessary to perform the cement life cycle 
inventory because there are other contradicting sources reporting a range of emission 
factors for Indian cement (0.6 to 1.0 tonnes CCVtonne cement). Actually this range is 
represenative of the how large companies and small companies generate a majority of the 
electricity on-site. However, the efficiency of production for the smaller companies is 
less than that of the larger companies, which seems to make the emission factors 
fluctuate. Other materials and transportation needed for concrete revealed that much of 
the emissions is arising from cement manufacturing. An MFA-LCA of cement in Rajkot 



218 



revealed per capita cement use in Rajkot is 0.15 tonnes/person more than Delhi but is still 
0.1 1 tonnes/person below that of a U.S. city like Denver. Final MFA-LCA calculations 
for pervious concrete and HVFA concrete mixtures showed at most a 21% and 47% 
reduction in emissions, respectively, compared to a conventional concrete used in Rajkot. 
6.1.2. Climate Adaptation: Pervious Concrete 

The purpose of this study was to determine environmental and structural 
properties of a pervious concrete demonstration in Rajkot, India. Changes in rain events 
can become an issue for stormwater solutions. Flooding are a concern for water quality, 
capacity and long-term durability of stormwater designs. The pervious concrete 
demonstration revealed that Rajkot materials were acceptable for making a pervious 
concrete mixture that provided adequate porosity and hydraulic. The typical porosity of 
15 to 25% and hydraulic conductivity above the impervious zone of approximately 0.15 
cm/s (0.06 in/s) were met. The pervious concrete also showed the water filtering 
capabilities and potential for reducing some polluting parameters such as nitrogen levels. 
However, similar to Hager's (2009) study pH levels do increase due to the lime present in 
concrete. Pervious concrete strength reached at least 6.9 MPa (1000 psi) which could be 
satisfactory for landscaping infrastructure. However, the long-term performance of 
strength was determined uncertain (on-average). Cubes only met design strength of 13.8 
MPa (2000 psi) once out of 4 batches; cylinders met design strength once out of 3 
batches. A strength relationship was deemed necessary for cross-country comparisons of 
strength since it is unclear which shape is more appropriate for representing strength for 
pervious concrete. However, due to the large spread in strength results [standard 



219 



deviations between 2.6 MPa (377 psi) and 3.8 MPa (561 psi)] in both cylinders and cubes 
a relationship was not determined at this time. 
6.1.3 Climate Adaptation: HVFA Concrete 

Although high temperatures have been addressed for making, placing, and curing 
concrete the subject of a high temperature environments along with effects of heated 
aggregate and amount of aggregate is not well addressed in research. This dissertation 
presented evidence that fly ash can be beneficial to concrete for maintaining lower 
internal concrete temperatures. The research also showed that when you have combined 
effects such as high temperatures, heated aggregate, and varying aggregate content and 
curing in hot weather conditions compressive strength results for ordinary portland 
cement (OPC) concrete mixtures will still perform better than high volume fly ash 
(HVFA) concrete mixtures designed for this study. Nevertheless, HVFA and OPC 
concrete mixtures remained above design strength for the 90 days of curing in extreme 
temperatures above 37.8°C (100°F). For permeability, on average OPC and HVFA 
mixtures showed decreases in permeability over time. If water cured both OPC and 
HVFA concrete mixtures demonstrate Low Permeability (1000 to 2000 coulombs). 
However if heat cured HVFA showed slightly less permeability than OPC mixtures, 
demonstrating Moderate Permeability (2000 to 4000 coulombs) by 90 days. For heat 
cured OPC mixtures permeability fell in to the high permeability range (above 4000 
coulombs). The percentage length change in water cured samples was very close to zero. 
HVFA heat cured samples did exhibit some shrinkage but about 0.02% less than OPC 
heat cured samples. Overall, length change measurements showed that HVFA concrete 
were about 50% lower than OPC concrete when heat cured. 



220 



Based on all tests results it was difficult to conclude whether HVFA concrete 
performed better than OPC concrete in terms of strength and durability. It was 
concluded, however, that these HVFA and OPC concrete mixtures could both be adjusted 
based on the (independent variables) cementitious material used, curing conditions used, 
aggregate content, and number of curing days allowed for the concrete. A multiple linear 
regression analysis was performed to model compressive strength, permeability, and 
length change (dependent variables) in order to determine which independent variables 
had a significant influence on the concrete mixtures adaptability to hot weather 
conditions. The multiple linear regression analysis showed that not all independent 
variables proved to be significant for the dependent variables measured. Strength, for 
example, was significantly affected from the variables cementitious material used, curing 
conditions used, aggregate content, and number of curing days. Heated or no heated 
aggregate did not have much of an effect on either of the mixtures for strength, 
permeability, and length change. It can be concluded that HVFA should be considered 
for extreme hot weather conditions but results also revealed that OPC concrete can be just 
as beneficial in these conditions. 
6.2 Contributions 
This study provided the following contributions to literature: 

• It was the first study to perform an MFA-LCA on-site mixed concrete specifically 
for India 

• It was the first study that provided a demonstration of pervious concrete to a city 
in Gujarat, India 



221 



• It was one of the first studies to provide a compressive strength a relationship 
between pervious concrete cube and cylinders when the mixture design involved a 
comparison of concrete materials. 

• It was one of the first studies to have a combination of heated aggregate, heat 
curing, cyclic temperatures, different aggregate content and long term curing in 
order to measure HVFA and OPC concrete's adaptability to hot weather 
conditions. 

• One of few studies with multiple linear regression for variable significance on 
HVFA and OPC in hot weather 

This study has also provided the following contributions directed towards practitioners: 

• Provided an MFA-LCA based tool for determining environmental impact of 
different on-site concrete mixed mixtures 

• Introduced the method and benefits of a pervious concrete system and suggested 
future use of system 

• Provided verification that two sources of Indian fly ash have beneficial properties 
towards achieving a concrete strength of 27.6 MPa (4000 psi) which is important 
to pavement and certain structural designs. 

• Provided verification that HVFA can perform above design strength in semi-arid 
to arid conditions and improve permeability and mitigate length change 

6.3 Recommendations and Future Research 

Based on the work described in this study many recommendations can be made to 
help improve the results. For example emissions estimates from cement and on-site 
concrete production can be represented by CO2 equivalents, the standard deviation of the 

222 



pervious concrete can be reduced through an increase in sample size and improvements 
made in the curing process, and sample size of the HVFA concrete tests can be increased 
by establishing a control and best performing concrete mixture in strength and durability 
separately. 

6.3.1 MFA-LCA Recommendations 

It was stated in Chapter 3 that other greenhouse gas emissions besides CO2 were 
not included due to unavailability of the data. CO2 equivalents can be estimated for India 
through use of nitrous oxide and methane data available from U.S. cement and concrete 
production or other country data. Additionally, Reiner's (2007) study included emissions 
from the use of water for mixing concrete. However, it was not clear whether all cities in 
India manually dig for water or use equipment for attaining water. A water emissions 
factor should be determined if necessary. Emissions were estimated for on-site mixed 
concrete, which is the dominant type of concrete used in concrete construction in India. 
However, ready mixed concrete operations are progressively increasing in India. 
Emissions from Indian ready mixed concrete are beneficial to estimate. Lastly, more 
research is recommended for determining the efficiency of captive power plants. 

6.3.2 Pervious Concrete Recommendations 
Image Analysis 

Orientation of aggregate and actual cross-sectional area of the pervious concrete 
surface can vary throughout the depth of a pervious concrete sample. Orientation of the 
aggregate can affect the mechanical properties of the pervious concrete such as 
compressive strength. In a study by Mahoub, Canler, Rathbone, Robl, and Davis (2009) 
a pervious concrete slab was tested for various properties. In particular the authors had 



223 



sliced the slab samples into various layers and identified the orientation of the aggregate. 
Observations proved the orientation was very different throughout the layers and these 
observations were identified as being important for protocols that help approximate field 
conditions of pervious concrete. Image analysis was recommended in this study as well. 
Identifying particle orientation in the future could be used to determine actual 
permeability (hydraulic conductivity). 
Weibull Statics 

Weibull statistics might be useful in determining specimen size effects on the 
compressive strength performance between a cylinder and a cube. For example, within a 
small sample or rather a small volume (i.e. cylinder with 100.5 in ) composite there is a 
finite probability that there will be a flaw sufficiently large enough to cause failure at a 
particular stress level. On a specimen made up of a large number of small volume 
elements (i.e. cube with 216 in 3 ) the probability of the existence of a serious flaw is much 
larger. Large specimens are therefore inherently weaker, and so will have lower ultimate 
strengths and give lower fatigue lives. Weibull statistics can be used to express survival 
probability in terms of specimen volume and stress or fatigue life. Weibull statistics, 
however, contradicts the information reported in Mindess, Young, and Darwin (2003) 
and Neville (1973) that suggests you most often get higher strengths with cubes. But 
another interpretation of Weibull statistics suggests that cylinders and cubes (100 .5 in 
and 216 in , respectively) do not sufficiently differ in volume as indicated by Weibull 
statistics (A large volume = a large number of smaller volume elements). More research 
into weibull statistics applicability to cylinders and cubes should be performed. 



224 



6.3.3 HVFA Concrete Recommendations 

Phase TV -Redesign for testing HVFA concrete in hot weather conditions 

The literature on concrete in urban areas exposed to hot weather conditions 
revealed that even in recent weather conditions concrete pavements tend to experience 
deterioration and detrimental effects due to hot weather. In 2012 several concrete 
pavements had buckled as a result of consecutive temperatures remaining above 32.2°C 
(90°F). Additionally literature suggested that not many studies have attempted to test 
concrete under a combination of hot weather conditions and adjust the mixture design to 
counteract the effect of the hot weather conditions. Therefore, in this research the hot 
weather conditions included curing concrete specimens under diurnal temperatures 
between 22.2°C (72°F) and above 37.8°C (100°F) and heating the aggregate to about 
65°C (149°F) just before including them in the mixture, to represent the aggregate being 
exposed to hot weather. To offset the effects of the hot weather, 50% fly ash (to replace 
cement) and varying the coarse aggregate content (of total aggregate) were used 
simultaneously in the mixtures. Fly ash is known to lower the internal heat during 
hydration and generally a higher use of coarse aggregate helps to reduce drying shrinkage 
problems. This information was discussed earlier in the chapter. 

The HVFA concrete mixtures were compared to OPC mixtures in terms of 
compressive strength, permeability, and length change during 90 days of curing (the 
recommendation for fly ash concrete is at least 56 days of curing). As discussed in the 
summary, the results revealed that overall OPC concrete had higher compressive strength 
throughout the duration of the testing. However, the HVFA concrete sample strengths 
remained above the design strength. The OPC concrete and the HVFA concrete samples 



225 



appeared to perform either similarly or the HVFA concrete performed slightly better than 
OPC concrete samples according to the results for permeability and length change. 

However, based on the limited amount of samples that were designated for the 
testing of compressive strength, permeability and length change a definitive conclusion 
could not be made between HVFA and OPC concrete under hot weather conditions. 
Even if the multiple linear regression analysis included all data without taking averages 
the standard deviation of the multiple linear regression equations, for example, did not 
encompass all variations in strength that was recorded during this study. Thus it is 
recommended by the author that the following testing plan be used to verify the 
performance of high volume fly ash concrete under a combination of hot weather 
conditions. 
Design of Experiment 

The experiment will involve two phases of testing. Phase I will be concerned 
with mostly later age concrete properties and results will be applied towards improving 
the results for multiple linear regression. Phase II will be focused on early age concrete 
properties and the results will be applied towards elastic potential energy, temperature 
profiles, and changes in strength. The course of testing will take approximately two years 
since 90 days of testing are needed for total curing time. Mixture designs will remain the 
same as listed in Table 5.11, however, the worst case scenario of hot weather conditions 
will be tested first. The order of testing will commence with two main mixtures 50% FA 
and OPC with 65% coarse aggregate content. Table 6.1 summarizes the order in which 
the mixtures will be tested. The base mixture design is shown in Table 6.2 which briefly 
mentions the cement, total aggregate content, water/cementitious ratio and the fly ash 



226 



replacement of the ordinary portland cement when required for the mixture. A water 
reducing admixture will be used as well and will be applied using the same dosage 
described in Table 5.11. 
Table 6.1 Order of Performing Mixtures 



Order of 
Testing 


Mixture 
Code 


Heated 

Aggregate 


1 


50FA65H 


Yes 


1 


OPC65H 


Yes 


1 


50FA65W 


Yes 


1 


OPC65W 


Yes 


2 


50FA65H 


No 


2 


OPC65H 


No 


2 


50FA65W 


No 


2 


OPC65W 


No 


3 


50FA55H 


Yes 


3 


OPC55H 


Yes 


3 


50FA55W 


Yes 


3 


OPC55W 


Yes 


4 


50FA55H 


No 


4 


OPC55H 


No 


4 


50FA55W 


No 


4 


OPC55W 


No 



Table 6.2 Base Mixture Design 



W/C+FA 


0.4 


FA replacement (%) 


50 


Design Compressive Strength 


27.6 MPa (4000 psi) 


Total Cementitious Content 
(kg/m 3 ) 


376.7 


Approximate Total Aggregate 
(kg/m 3 ) 


1800 



1 kg/m 3 =1.68554 lb/yd 3 
Phase I testing is summarized in Table 6.3. Phase II testing is summarized in Table 6.4. 
Figure 6.1 shows a sample of what the testing schedule could look like for Phase I 
through Phase II. Both water and heat curing setup of equipment and tanks will remain 



227 



the same as was used in this dissertation. However, a humidity sensor is recommended 
for the heat curing tank. The humidity sensor can be connected to the datalogger. 



Table 6.3 Phase I Testing 


Summary for Each Mixture 












Hardened 
Property Tests 


Total 
Beams 


Total 
Cylinders 


Testing Days 


1 


2 


3 


4 


5 


6 


7 


14 


28 


56 


90 


Length Change 


3 




X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


Permeability 




6 (2 per 
each test) 


















X 


X 


X 


Compressive 
Strength 




18 (6 per 
each test) 


















X 


X 


X 



Table 6.4 P 


hase II Testing 


Summary for Each Mixture 












Hardened 
Property Tests 


Total 
Beams 


Total 
Cylinders 


Testing Days 


1 


2 


3 


4 


5 


6 


7 


14 


28 


56 


90 


Length Change 


3 




X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


Internal 
Temperature 




5 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


Modulus of 

Elasticity and 

Compressive 

Strength 




18 (3 per 
each test) 


X 




X 








X 




X 


X 


X 



Hot Weather 
Testing HVFA 
Schedule 

Phase I - Testing 1 

Phase I - Testing 2 

Phase I - Testing 3 

Phase I -Testing 4 

Phase II - Testing 1 

Phase II - Testing 2 

Phase II - Testing 3 

Phase II - Testing 4 



y o o o o o o 



y v ,v \ 



y v ,v n 



>■ 






^ \ N 



y y 



*P ^ ^ o° v <^ # # ^ *P o* <f # 




Figure 6.1 Sample Schedule for Competing Phase I-II Testing 

Phase I Testing: Improving Multiple Linear Regression Analysis 

In the multiple linear regression analyses the results suggested that heated 
aggregate may not be of statistical significance and have no influence on the strength, 
permeability, and length change results. Nevertheless, since not enough samples were 



228 



used to make this conclusion the condition of using heated aggregate in the mixture is 
still recommended for the testing described in this section. 

Later age testing results are preferred for the multiple linear regression analysis 
for a couple of reasons. (1) Strength testing for both the HVFA and OPC concrete begin 
to reach a stable gain in strength by 28 days of curing. Before 28 days the rate of strength 
gain constantly changes (especially between and 7 days of strength) as can be seen in 
Figures 5.17a and 5.17b. Therefore, it is expected the major changes in strength gain 
should be mainly be influenced by the curing conditions and if there were heated 
aggregate in the mixtures as is seen in Figures 5.18a and 5.18b, (2) Permeability testing 
is more commonly tested at or after 28 days of curing. 

The samples tested in Phase I will be included with the results that were recorded 
during the completion of this dissertation thus increasing the number of samples for the 
multiple linear regression analysis. After combining the sample results between this 
dissertation and Phase I then compressive strength will have a total of at least 8 samples 
per day, permeability will have 4 samples per day, and length change will have 4 samples 
per day for each mixture. 

Phase II Testing: Early age data to model length change and relate thermal evolution 
effects to fracture patterns. 

Phase II testing will aim towards collecting more data regarding length change, 
modulus of elasticity, and internal temperatures of the concrete while curing. In addition 
to length change being useful in the multiple linear regression analysis, length change 
measurements will be beneficial towards developing a finite element model of a 
pavement slab that is exposed to hot weather conditions. The ultimate goal of the finite 



229 



element model would be to simulate buckling of the pavement slab. The modulus of 
elasticity will be used to determine the potential energy that can develop in heat and 
water cured concrete samples when subjected to compressive loading. Measuring 
internal temperatures will be useful in confirming whether HVFA concrete can maintain 
at least lower temperatures than the OPC concrete. Lower concrete temperatures can 
reduce thermal stresses, drying shrinkage cracking, and permeability, and help preserve 
long term concrete strengths. In the next few paragraphs the method of either using the 
data or taking measurements for length change, modulus of elasticity, and internal 
temperatures will be discussed further. 
Length Change Modeling 

The factors that contribute to cracking and failure of early age concrete may be 
due to temperature, creep, and shrinkage. Creep is a deformation that occurs in the 
concrete over time and is dependent on the loading imposed on the concrete. Whether 
the concrete is loaded or not, shrinkage results from the chemical and physical changes 
during the hydration process and will be affected by the surrounding environment. The 
occurrence of creep and shrinkage can be linked to buckling. As part of this phase of 
testing it is implied that knowing the length change of concrete exposed to hot weather 
conditions over time will be useful in constructing a finite element model. The idea is to 
model a concrete pavement slab built up of rectangular shapes similar in size to those 
beams tested in lab for length change. Figure 6.2 shows a simple example of what the 
finite element mesh would look like. 



230 




11.6" (294.6 mm) 



Figure 6.2 Example of Finite Element Mesh and a Close-Up of a Single Element 
Based on Dimensions of the Length Change Beam Made in Lab 

Modulus of Elasticity and Potential Energy 

Early age (1-14 days of curing) strength tests and modulus of elasticity tests were 

originally conducted in the research. The early age strength test showed how heat cured 

concrete (including those samples with heated aggregate) gained strength at a slightly 

higher rate than water (ideally) cured concrete (Refer to Figures 5.17 and 5.19). Within 

this dissertation the early age modulus of elasticity data was only used to recognize the 

pattern of the concrete's stiffening process and the information gathered on modulus of 

elasticity was also going to be used to determine the potential energy stored in heat cured 

concrete versus water cured concrete while under compressive loading. Under Phase II 

testing the determination of the elastic potential energy from compressive loads will be 

explored further. The elastic potential energy became of interest for this research because 

a comparison between HVFA and OPC concrete, when testing for compressive strength, 

revealed different failures. For example, heat cured OPC concrete samples tend to break 

or fracture explosively versus and the subtle or softer failures that the heat cured HVFA 

concrete demonstrated. Elastic potential energy is the work done to deform object being 

loaded but can return to its original shape if the loading is released. Elastic potential 

energy can be studied rather than fracture energy because during testing of modulus of 

231 



elasticity we can determine the energy stored in the sample from a stress versus strain 
graph. Figure 6.3 demonstrates how the stress of an OPC water and heat cured sample 
are plotted against the strain of the concrete samples. The area under the approximate 
straight lines represents the elastic potential energy. The difference in energy is the 
shaded area shown in Figure 6.3. In this case it appears that more energy is stored in a 
water cured sample versus a heat cured sample of OPC concrete, therefore, failure in the 
water cured sample might be assumed to be more pronounced than the heat cured sample. 



3500 



3000 



2500 



on 
g2000 

H 
95 

01 

I 1500 

■- 

p. 

E 

3 iooo 



500 




20 



15 



10 



O 

o 



■a 
i 

re 



n 

I 

era 



-0 



0.0001 



0.0002 



0.0003 


0.0004 


Strain 




W90 Day - 


- H90 Day 



0.0005 




0.0006 



Figure 6.3 Difference between Elastic Potential Energy of Water Cured and Heat 
Cured OPC Concrete Sample after 90 Days of Curing 

Internal Temperatures 

Early age data is especially important when measuring the internal temperature of 

the concrete as it is curing. As was seen in Figures 5.10 and 5.11 the temperature profiles 

showed that HVFA concrete can successfully maintain the internal temperatures of the 

232 



concrete at about 5% to 11% lower than OPC concrete internal temperatures (depending 
on the temperature of the surrounding environment). This difference in temperature is 
best measured during the first 3 days of curing when the exothermic reaction occurs 
during the hydration process. Having a lower internal temperature during hydration can 
be beneficial towards concrete cured in hot weather conditions. Lower internal 
temperatures of concrete can possibly help moderate evaporation and thermal cracking 
during the curing process. Originally, during this study, temperatures were being 
recorded for each mixture throughout the curing process; however, the thermocouples 
placed within the concrete would sometimes be damaged and stop recording 
temperatures. It was not clear if the damage occurring was due to a reaction between the 
thermocouple wires and the cement paste or if the wires were being crushed by the 
hardening of the cement paste. Another method, compared to the one used in this 
dissertation is being recommended. 
Alternative Method 

Instead of using a liquid tape to protect the ends of the twisted wires an epoxy will 
be used that can harden to about a 6.35 mm (1/4 inch) thick once dried. The 
epoxy can be purchased from any hardware store. Additionally thicker 
thermocouple wires should be used as well. The diameter of the wire used in this 
dissertation was a AWG 24. The same Type J thermocouple wire with a diameter 
of AWG 20 shall be used instead. The placement of the wire will be similar to 
that described in the dissertation. Thus, the thermocouple wire is placed inside 
the cylinder mold, half way the length of the cylinder mold while attached to a 



233 



wood dowel. Figure 6.4 provides a schematic showing the placement of the 
dowel and thermocouple in the concrete cylinder. 



to data,, 
acquisition system 



in (204 mm) 



-Type J Thermocouple 



-Dowel 



..-^ Concrete Cylinder 

— 4 in (102mm)^ 

Figure 6.4 Schematic of Placement of the Thermocouple in Concrete Cylinder 

Summary of Experimental Plan 

An experimental plan has been outlined for quantifying the effects of hot weather 
conditions (heat curing in diurnal temperatures, and heated aggregate) and is compared to 
the effects of ideal water curing. The mechanical properties (compressive strength, 
length change and permeability) of HVFA and OPC concrete, while being exposed to the 
hot weather conditions, would be compared. The curing conditions are intended to 
simulate changes in the climate that cause temperatures to remain above 37.8°C (100°F) 
for long periods of time. This experimental outline is a result of the outcomes of this 
dissertation. The outcomes revealed that although HVFA concrete had lower strengths 
than the OPC concrete samples the HVFA never fell below design strength and HVFA 
had comparable or slightly better resistance to permeability and length change compared 
to the OPC concrete samples. However, as part of this dissertation the results are only 
preliminary and cannot be used to make definite conclusions until supported with more 
data. Thus this experimental outline presents two major goals through two phases of 

234 



testing. Phase I is meant to accomplish increasing the number of samples to replicate the 
results presented in this dissertation. Phase I and II testing together have the purpose of 
extending the use of the data in models. The length change is intended to be modeled as 
shrinkage experienced by pavements exposed to hot weather conditions, applying 
compressive strength will have the purpose of quantifying differences in stored energy by 
HVFA and OPC concrete which should lead to an understanding of the how either of the 
concrete mixtures could fail while in service. And finally, modeling of the temperature 
profile can be used to verify lower curing temperatures for HVFA concrete compared to 
OPC concrete while placed in hot weather conditions. 
6.4 Final Remarks Regarding Sustainability 

This study was developed as part of the advancement of interdisciplinary research 
and sustainability. In order to have completed this study it involved the comprehension 
of other disciplines, culture, social behavior that allowed for collaboration with other 
individuals and organizations that are directly involved with the development of 
infrastructure. It is the hope of the author that the audience developed the 
understanding that although emissions and other impacts on the environment are most 
likely arising from the use of materials, energy, equipment, etc., society is still highly 
dependent on each of these things. The goal of this dissertation was meant to improve 
upon the current knowledge of materials that we use on a daily basis, such as concrete, so 
as to find ways to reduce impacts on society and the environment with current 
infrastructure technology. 



235 



References 

Alimchandani, C. R. (2007). The Indian Experience of Ready Mixed Concrete (RMC). 
Proceedings of the 32" Conference on Our World in Concrete & Structures, Singapore 
28-29 August 2007. 

Ambuja Cements Ltd. (2010). Ambuja Cement Annual Report 2010. Retrieved in 2012 
from http://www.ambuiacement.com/ 

American Coal Ash Association (ACAA). (n.d.). Coal ash fact sheet. Retrieved from 
http://www.acaa-usa.org/associations/8003/files/Coal Ash Fact Sheet-ACAA-ACC.pdf 

American Coal Ash Association (ACAA). (2002). The benefits of using fly ash in 
concrete. Retrieved from http://www.acaa- 
usa.org/associations/8003/files/The Benefits of FA in Concrete-CIRCA.pdf 

American Coal Ash Association. (2003). Fly ash facts for highway engineers (FHWA-IF- 
03-019). Washington, D.C. 

American Coal Ash Association. (2010). 2009 Coal Combustion Product (CCP) 
Production and Use Survey Report. Retrieved from http://www.acaa- 
usa.org/associations/8003/files/2009 Production and Use Survey Goss FINAL 11031 
O.pdf 

American Coal Ash Association. (2009). 2008 Coal Combustion Product (CCP) 
Production and Use Survey Report. Retrieved from http://www.acaa-usa.org/ 
American Concrete Institute (ACI). (2010). Guide to Hot Weather Concreting (ACI 
305R-10) 

American Concrete Institute (ACI). (1996). Cement and concrete terminology (ACI 
116R) 

American Concrete Institute (ACI). (1996). Use of fly ash in concrete (ACI 232. 2R) 

American Society of Civil Engineers (ASCE). (2009). 2009 Report card for America 's 
infrastructure. Retrieved from 
http://www.infrastructurereportcard.org/sites/default/files/RC2009 full report.pdf 

ASTM C31-06: Practice for Making and Curing Concrete Test Specimens in the Field 

ASTM C39M-05: Test Mehtod for Compressive Strength of Cylindrical Concrete 
Specimens 

ASTM C94-07: Specification for Ready-Mixed Concrete 



236 



ASTM C138-07: Test Method for Density (Unit Weight), Yield, and Air Content 
(Gravimetric) of Concrete 

ASTM C143-05a: Test Method ofr Slump of Hydraulic-Cement Concrete 

ASTM C 172-07: Practice for Sampling Freshly Mixed Concrete 

ASTM C231-04: Test Method for Air Content of Freshly Mixed Concrete by the Pressure 
Method 

ASTM C 1064-05: Test method for Temperature of Freshly Mixed Hydraulic-Cement 
Concrete 

Atis, C. D. (2002). Heat evolution of high volume fly ash concrete. Cement and Concrete 
Research 32, 751-756 

BCS Incorporated. (2008). Waste heat recovery: Technology and opportunities in U.S. 
Industry. Contract report under U.S. Department of Energy Industrial Technologies 
Program. Retrieved in 2012 from 

https://wwwl.eere.energv.gov/manufacturing/intensiveprocesses/pdfs/waste heat recove 
ry.pdf 

Bentz, D. P., Peltz, M. A., Duran-Herrera, A., Valdez, P., & Juarez, C. A. (201 1). 
Thermal properties of high-volume fly ash mortars and concretes, Journal of Building 
Physics, 34(3), 263-275. doi:10.1177/1744259110376613 

Boyle, C, Mudd, G., Mihelcic, J. R., Anastas, P., Collins, T., Culligan, P., Edwards, M., 
Gabe, J., Gallagher, P., Handy, S., Kao, J. J., Krumdieck, S., Lyles, L. D., Mason, I., 
Mcdowall, R., Pearce, A., Riedy, C, Russell, J., Schnoor, J. L., Trotz, M., Venables, R., 
Zimmerman, J. B., Fuchs, V., Miller, S., Page, S., & Reeder-Emery, K. (2010). 
Delivering Sustainable Infrastructure that Supports the Urban Built Environment. 
Environmental Science and Technology, 44 (13), 4836-4840. 

Bureau of Energy Efficiency (2012). Fuels and Combustion, Energy Efficiency in 

Thermal Units. Retrieved in 2012 from 

http://www.beeindia.in/content.php ?page=energy managers auditors/ema.php?id=4 

Caulkins, J., Kney, A., Suleiman, M. T., Weidner, A. (2010). "Removal of heavy metals 
using pervious concrete material". World Environmental and Water Resources Congress 
2010: Challenges of Change-Proceedings of the World Environmental and Water 
Resources Congress 2010. 

Central Development Mechanism (CDM), Bhat, P. (2006). CDM in India, challenges and 
success [Powerpoint slides] . Retreived from 
http://www.epco.in/pdf/CDM India Challenges.pdf 



237 



Cement Manufacturers' Association (CMA). (2010). Annual Report 2009-2010. 
Retrieved in 2012 from http://www.cmaindia.org/portal/index/index.aspx 

Cement Manufacturers' Association (CMA). (2010). Cement Manufacturers ' Association 
of India. Retrieved from http://www.cmaindia.org/portal/index/index.aspx 

Cement Manufacturers' Association (CMA). (2010). Cement Statistics 2010. 

Center for Clean Air Policy (CCAP) & The Energy And Resources Institute (TERI). 
(2006). Greenhouse gas mitigation in India: Scenarios and opportunities through 2031. 
New Delhi, India 

Central Electricity Authority (CEA). (2009). CO2 Basline Database for the Indian Power 
Sector. Government of India Ministry of Power-Central Electricity Authority 

Central Pollution Control Board (CPCB). (2009). Comprehensive Industry Document 
Stone Crushers. Prepared by Parivesh Bhawan Retrieved from www.cpcb.nic.in 

Chavez, A., Ramaswami, A., Dwarakanath, N., Guru, R., & Kumar, E. (2012). 
Implementing Trans-Boundary Infrastructure-Based Greenhouse Gas Accounting for 
Delhi, India. Journal of Industrial Ecology, 16 (6), 814-828 

Colorado Ready Mixed Concrete Association [CRMCA]. (2009). Specifiers guide for 
pervious concrete pavement design. Retrieved from 
http://www.udfcd.org/downloads/down sw bmp.htm 

Confederation of Indian Industry (CII), Natural Resources Canada (NRC), Ambuja 
Cement (2004). High Volume Fly Ash Concrete Technology Concrete Road Project 
Saurashtra University Rajkot, Gujarat. 

Crow, J. M. (2008). The concrete conundrum. Chemistry World, 5(3), 62-66 

Davis, R. E., Carlson, R. W., Kelly, J. W., & Davis, H. E. (1937). Properties of cements 
and concretes containing fly ash. Journal of the American Concrete Institute, 33, 577-612 

Delatte, N., Mrkajic, A., and Miller, D. I. (2009). Field and Laboratory Evaluation of 
Pervious Concrete Pavements. Journal of the Transportation Research Board, 2113, 
132-139. 

Delatte, N., Miller, D., & Mrkajic, A. (2007). Portland Cement Pervious Concrete 
Pavement: Field Performance Investigation of Parking Lot and Roadway Pavements. 
RMC Research & Education Foundation. 

del Viso, J. R., Carmona, J. R, Ruiz, G. (2007). Shape and size Effects on the 
compressive strength of high- strength concrete. Cement and Concrete Research, 38, 386- 
395. 



238 



Department Related Parliamentary Standing Committee on Commerce (DRPSCC). 
(201 1). Performance of( 
Rajya Sabha New Delhi. 



(201 1). Performance of Cement Industry, 95 l report presented to Parliament of India 



Dhadse, S., Kumari P., Bhagia L. J. (2008). Fly ash characterization, utilization and 
government initiatives in India- A review. Journal of Scientific & Industrial Research. 
67,11-18 

Dutta, M., Mukherjee, S. (2010). An outlook into energy consumption in large scale 
industries in India: The cases of steel, aluminum, and cement. Journal of Energy Policy, 
35,7286-7298 

Environmental Information System (ENVIS) Centre, Department of Forests, Ecology, 
and Environment (DFEE), Government of Karnataka (2007). Utility bonanza from dust. 
Parisa ENVIS Newsletter. 2(6) 

Environmental Protection Agency (EPA). (2008). Direct emissions from stationary 
combustion sources. Climate leaders greenhouse gas inventory protocol core module 
guidance. Retreived from 

http://www.epa.gov/climateleadership/documents/resources/stationarvcombustionguidanc 
e.pdf 

Ernst & Young. (201 1). " Cement" -ing growth. India: Ernst & Young Pvt. Ltd. 

European Coal Combustion Products Association (ECOBA). (n.d.). What are CCPs. 
Retrieved from http://www.ecoba.com/ecobaccps.html 

Federal Highway Administration (FHWA). (2012). User Guidelines for Waste and 
Byproduct Materials in Pavement Construction. Retrieved from 
http://www.fhwa.dot.gov/publications/research/infrastructure/structures/97148/app2.cfm 

Federal Highway Administration (FHWA). (1999). Fly Ash. Retrieved from 
http://www.fhwa.dot.gov/infrastructure/materialsgrp/flyash.htm 

Federation of Indian Mineral Industries (FEVII). (2009). Statistics: Production of all ores 
and minerals (excluding atomic & minor minerals). Retrieved from 
http://www.fedmin.com 

Gani, M. S. J. (1997). Cement and Concrete. London, UK: Chapman & Hall 

Garg A., Shukla P.R., Kapshe M. (2006). The sectoral trends of multigas emissions 
inventory of India. Atmospheric Environment, 40 (24), pp. 4608-4620. 

Ghafoori, N., & Dutta, S. (1995). Development of no-fines concrete pavement 
applications. Journal of Transportation Engineering, 121(3), 283-288. 



239 



Giaccio, G. M. & Malhotra, V. M. (1988). Concrete incorporating high-volumes of 
ASTM Class F fly ash [Abstract]. ASTM Journal of Cement, Concrete, and Aggregates, 
10 (2), 88-95. 

Gordon, D., Kshemendranath, P., (1999). Setting up of ready mixed concrete industry in 
India: The challenges and opportunities. In Dhir, R. K., and Limbachiya, M. C. (Eds.), 
Utilizing ready-mixed concrete and mortar (21 1-220). London: Thomas Telford 
Publishing 

Grasim Industries Limited. (2008). Cementing Relations Sustainability Report 07-08. 
Retrieved from http://www.grasim.com/sustainability/sustainability.htm 

Hager, A. S. (2009). Sustainable design of pervious concrete pavements. Ph.D. 
dissertation, University of Colorado at Denver, United States — Colorado. Retrieved 
January 30, 201 1, from Dissertations & Theses @ University of Colorado 
System.(Publication No. AAT 3360870). 

Hathi Cement. (201 1). 53 rd Annual Report 2010-11. Retrieved in 2012 from 
http://www.hathi-sidheecements.com/hathi site/index.html 

Hayter, A. J. (1996). Probability and Statistics. Boston, Massachusetts: PWS Publishing 
Company 

Headwaters Resources. (2005). HVFAfor Concrete Structures. Bulletin number 23 

Hendriks, C. A., Worrell, E., de Jager, D., Blok, K., Riemer, P. (2004). Emission 
reduction of greenhouse gases from the cement industry. Proceedings from the 
Greenhouse gas technologies conference, Vancouver, Canada 2004 

Intergovernmental Panel on Climate Change (IPCC). (2007). The Physical Science Basis, 
IPCC Fourth Assessment Report. 

Intergovernmental Panel on Climate Change (IPCC). (2007). Summary for Policy 
Makers. Synthesis Report, IPCC Fourth Assessment Report. 

Intergovernmetnal Panel on Climate Change (IPCC). (2006). Industrial Processes and 
Product Use. 2006 IPCC Guidelines for National Greenhouse Gas Inventories Retrieved 
from http://www.ipcc-nggip.iges.or.jp/public/2006gl/vol3.html 

Intergovernmetnal Panel on Climate Change (IPCC). (1996). Industrial Processes. 
Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories Retrieved from 
http://www.ipcc-nggip.iges.or.jp/public/gl/invs5.html 

International Energy Agency. (2011). Energy transition for industry: India and the global 
context. Prepared by Trudeau, N., Tarn, C, Graczyk, D., and Taylor, P. Paris, France 



240 



Kennedy C, Ramaswami A., Carney S., and Dhakal S. (2009). Greenhouse Gas Emission 
Baselines for Global Cities and Metropolitan Regions. Proceedings of the 5th Urban 
Research Symposium, Marseille, France, June 28-30, 2009. 

Khoury, G. A. (2006). Strain of heated concrete during two thermal cycles. Part 3: 
isolation of strain components and strain model development. Magazine of Concrete 
Research, 58(1), 421-435 

Kishore, K. (2012). Understanding nominal and design mixes. [Web long comment]. 
Retrieved from www.engineeringcivil.com/understanding-nominal-anddesign-mixes.html 

KPMG International. (2012). The great global infrastructure opportunity: Global 
construction survey 2012. Retreived 2012 from 

http://www.kpmg.com/EE/et/IssuesAndInsights/ArticlesPublications/Documents/GCS20 
12-web.pdf 

Kumar, P., Kaushik, S. K. (2003). Some trends in the use of concrete: Indian scenario. 
The Indian Concrete Journal, 1503-1508 
Lafarge. (2012). Site mix vs. ready mix. Retrieved from 
http://www.lafarge.in/wps/portal/in/3 A 2 2-Site mix vs ready mix 

Lorenzoni, I., Nicholson-Cole, S., Whitmarsh, L. (2007). Barriers perceived to engaging 
with climate change among the UK public and their policy implications. Journal of 
Global Environmental Change, 17, 445-459. 

Luck, J. D., Workman, S. R., Coyne, M. S., & Higgins, S. F. (2009). Consequences of 
manure filtration through pervious concrete during simulated rainfall events. Biosy stems 
Engineering, 102, 417-423 

Mahboub, K.C., Canler, J., Rathbone, R., Robl, T., & Davis, B. (2009). Pervious 
Concrete: Compaction and Aggregate Gradation. AC I Materials Journal, 106 (6), 523- 
528. 

Majersky, G. M. (2008). Filtration and Live Span Analysis of a Pervious Concrete Filter. 
University of Colorado Denver, Environmental Engineering. Denver, Colorado: 
University of Colorado Denver MS Thesis. 

Malhotra, V. M., & Mehta, P. K. (2008). High Performance, High Volume Fly Ash 
Concrete (3 edition). Ottawa, Canada: Supplementary Cement Materials for Sustainable 
Development Inc. 

Marceau, M. L., Nisbet, M. A., VanGeem, M. G. (2010). Life Cycle Inventory of 
Portland Cement Manufacture. Skokie, Illinois: Portland Cement Association. 



241 



McCarthy, M. P., Best M. J., Betts, R. A. (2010). Climate change in cities due to global 
warming and urban effects. Geophysical Research Letters, 37, 1-5. 
doi: 10. 1029/20 10GL042845 

McKinsey & Company. (2010). Building India Transforming the nation 's logistics 
infrastructure. Retrieved from www.mckinsey.com 

Mehrotra, S., Natenzon, C. E., Omojola, A., Folorunsho, R., Gilbride, J., Rosenzweig, C. 
(2009). Framework for City Climate Risk Assessment, Proceedings of the Fifth Urban 
Research Sysmposium. 

Mehta, P. K. (2002). Performance of high-volume fly ash concrete in hot weather, In V. 
M. Malhotra (Ed.). Innovations in design with emphasis on seismic, wind, and 
environmental loading; quality control and innovations in materials/hot -weather 
concreting (pp. 47-52). Farmington Hills, Michigan: American Concrete Institute. 

Mehta, P. K., & Monteiro, P. J. M. (2006). Concrete micro structure, properties, and 
materials (3 rd ed.). New York, NY: McGraw-Hill. 

Mindess, S., Young, J. F., Darwin, D. (2003). Concrete (2 nd ed.). Upper Saddle River, 
NJ: Pearson Education, Inc. 

Ministry of Environment and Forests (MoEF). (2010). India Greenhouse Gas Emissions 
2007. Retrieved from http://moef.nic.in/downloads/public- 
information/Report INCCA.pdf 

Mirolli, M. (2005). The Kalina Cycle for Cement Kiln Waste Heat Recovery Power 
Plants. Cement Industry Technical Conference, 2005. May 15-20, 2005. 

Morrison, C. L. (201 1). The Stormwater Solution. Retrieved from 
http://www.cement.org/buildings/sustainable pervious.asp 

National Ready Mixed Concrete Association (NRMCA). (2004). CIP 38 - Pervious 
Concrete. Concrete in Practice. Retrieved from 
http://www.nrmca.org/aboutconcrete/cips/38p.pdf 

National Ready Mixed Concrete Association. (2007). Pervious Concrete Contractor 
Certification. Silver Spring, MD, USA: National Ready Mixed Concrete Association. 

Neptune, A. I. & Putman, B. J. (2010). Effect of Aggregate Size and Gradation on 
Pervious Concrete Mixtures 

Neville, A. M. (1973). Properties of Concrete. New York, NY: John Wiley and Sons Inc. 

Parikh J., Duttabiswas, C. R., Nirbhay, S., Kumar, V., Thapa, B., Dean, B., Ojha, S., 
Goel, R., Mathur, V., Khare, A., & Bhan, N. (2009). Analysis ofGHG Emissions for 



242 



Major Sectors in India: Opportunities and Strategies for Mitigation, Center for Clean Air 
Policy, Integrated Research and Action for Development, ICF International 

Parikh, J, Sharma, B. D., Kumar, V., Vimal, A., Integrated Research and Action for 
Development (IRADe). (2009). Indian Cement Industry. Proceedings of the Low Carbon 
Technologies; Implementation and Policy Issues, Power point presentation, New Delhi, 
Delhi, March 19, 2009. 

Portland Cement Association (PCA). (2002). Design and Control of Concrete Mixtures 
(14 l edition). Skokie, Illinois: Portland Cement Association 

Rajkot Municipal Corporation. (2006). Rajkot City Development Plan. (Unpublished city 
report). Rajkot, India. 

Ramachandra, T. V., Shwetmala, (2009). Emissions from India's transport sector: 
Statewise synthesis. Atmospheric Environment, 43, 5510-5517. 

Ramakrishnan, N. (2012, June, 12). Cement companies see savings in captive power 
plants. The Hindu Business Line. Retrieved from 
http://www.thehindubusinessline.com/industry-and- 
economy/article3 520286. ece?ref = wl opinion 

Ramaswami, A., Hillman, T., Janson, B., Reiner, M., Thomas, G., (2008). A demand 
centered, hybrid life-cycle methodology for city-scale greenhouse gas inventories. 
Environmental Science and Technology, 42(17), pp. 6455-6461. DOI: 
10.1021/es702992q 

Ramaswami, A. (2005). Engineering Sustainable Urban Infrastructure. In M. A. Abraham 
(Ed.), Sustainability Science and Engineering: Defining Principles (pp. 41 1-434). 
Amsterdam: Elsevier B. V. 

Ray, B. K. and Reddy, B. S. (2009). Decomposition of Energy Consumption and Energy 
Intensity in Indian Manufacturing Industries. Energy for Sustainable Development, 14, 

35-47. 

Ravina, D. (1981). Fly ash concrete under hot weather conditions, Proceedings of the 
Materials Research Society Annual Meeting on Effects ofFlyash Incorporation in 
Cement and Concrete, Boston Massachusetts, 172-183. 

Ready Mix Concrete Manufacturer's Association (RMCMA). (2008). Certification of 
Ready-Mixed Concrete Production Facilities Checklist. Quality Manual Part-1. Mumbai: 
RMCMA. Retrieved from http://www.rmcmaindia.org/ 

Reiner, M. (2007). Technology, environment, resource and policy assessment of 
sustainable concrete in urban infrastructure. Ph.D. dissertation, University of Colorado at 



243 



Denver, United States — Colorado. Retrieved January 26, 2011, from Dissertations & 
Theses @ University of Colorado System. (Publication No. AAT 3273499). 

Schindler, A. K. (2004). Effect of Temperature on Hydration of Cementitious Materials. 
ACI Materials Journal, 101 (1), 72-81 

Schumacher, K., Sathaye, J. (1999). India's cement industry: Productivity, energy 
efficiency, and carbon emissions. (Environmental Energy Technologies Division LBNL- 
41842) Berkeley, CA: Ernest Orlando Lawrence Berkeley National Laboratory 

Senthil, S., and Santhakumar, A. R. (2005). Quantification of temperature rise during hot 
weather concreting, Proceedings of the 6 International Congress Conference on Global 
Construction: Ultimate Concrete Opportunities Young Researcher 's Forum, University 
of Dundee, Scotland, 59-66. 

Shukla, P. R., Biswas, D., Nag, T., Yajnik, A., Heller, T., & Victor, D. G. (2004). Captive 
Power Plants :Case Study of Gujarat India. A working paper from the Program on 
Energy and Sustainable Development at the Center for Environmental Science and 
Policy. Retrieved 2012 from the website http://cesp.stanford.edu/pesd 

Sidhee Cement. (2011). 37 th Annual Report 2010-11. Retrieved in 2012 from 
http://www.hathi-sidheecements.com/sidhee site/index .html 

Siemens, GlobeScan, MRC McLean Hazel. (2007). Megacity challenges: a stakeholder 
perspective. Munich: Seimens AG Retrieved 2012 from 

http://www.siemens.com/entry/cc/features/urbanization development/all/en/pdf/study m 
egacities en.pdf 

Sisavath, S., Jing, X., and Zimmerman, R. W. (2001). Laminar Flow Through 
Irregrularly-Shaped Pores in Sedimentary Rocks. Transport in Porous Media, 45, 41-62 

Solis, A. V., Durham S. A., Rens, K. L., & Ramaswami, A. (2010). Sustainable concrete 
for the urban environment: A proposal to increase fly ash use in concrete, Proceedings of 
the ASCE Green Streets and Highways 2010 Conference: An Interactive Conference on 
the State of the Art and How to Achieve Sustainable Outcomes, 389 (33), 401-407 doi: 
10.1061/41148(389)33 

Solis, A. V., Durham S. A., & Ramaswami, A. (2012). Providing Stormwater 
Management Solutions in Rajkot India: A Pervious Concrete System Demonstration. The 
International Journal of the Constructed Environment, 2 (3), 135-154. 

Tarr, J. A. (1984). The evolution of the urban infrastructure in the nineteenth and 
twentieth centuries, In R. Hanson (Ed.), Perspectives on urban infrastructure (pp. 4-66). 
Washington, D.C.: National Academy Press. 



244 



Tennis, P. D., Leming, M. L., & Akers, D. J. (2004). Pervious Concrete Pavements. 
Skokie, Illinois: Portland Cement Association. 

The Indian Concrete Journal (2004). Can HVFAC technology be adopted for site-mixed 
concrete? 

UltraTech. (2012). Milestones. Retreived from 
http://www.ultratechcement.com/milestones.php 

UltraTech Cement Limited. (201 1). UltraTech Cement Limited Annual Report 2009- 
2010. Retrieved in 2012 from http://www.ultratechcement.com/ 

United Nations (UN). (2007). World Urbanization Prospects The 2007 Revision 
Highlights. Retrieved from 
http://www.un.org/esa/population/publications/wup2007/2007WUP Highlights web.pdf 

United Nations. (2009). World Urbanization Prospects The 2009 Revision Highlights. 
Retrieved from http://esa.un.org/unpd/wup/Documents/WUP2009 Highlights Final.pdf 

United Nations. (2010). UN 2010 among the deadliest years for disasters, urges better 
preparedness. Retrieved from 
http://www.un.org/apps/news/story.asp?NewsID=37357&Cr=disaster+reduction&Crl 

United Nations. (2010). Population Division of the Department of Economic and Social 
Affairs of the United Nations Secretariat. World Population Prospects: The 2010 
Revision Retrieved from http : //es a. un . org/unpd/wpp/index . htm 

U.S. Department of Energy (2003). Energy and emission reduction opportunities for the 
cement industry. Prepared by Choate, W. T. Columbia, MD: BCS Inc. 

U.S. Energy Information Administration (EIA). (2010). Environment: energy-related 
emissions data & environmental analyses. Retrieved from 
http://www.eia.doe.gov/environment.html 

U.S. Environmental Protection Agency. (2005). Using coal ash in highway construction: 
A guide to benefits and impacts. EPA 530-K-05-002. 

U.S. Geological Survey (USGS). (2012). Mineral Commodity Summaries. Retrieved from 
http ://minerals . usgs . go v/minerals/pubs/commodity/cement/ 

U.S. Geological Survey (USGS). (2011). Mineral Commodity Summaries. Retrieved from 
http ://minerals . usgs . go v/minerals/pubs/commodity/cement/ 

U.S. Geological Survey (USGS). (2010). Minerals Yearbook. Retrieved from 
http ://minerals . usgs . go v/minerals/pubs/commodity/cement/ 



245 



Venkatarama Reddy, B. V., & Jagadish, K. S. (2003). Embodied energy of common and 
alternative building materials and technologies. Energy and Buildings, 35, 129-137 

Wang, X., Nguyen, M., Stewart, M. G., Syme, M., Leitch, A. (2010). Analysis of Climate 
Change Impacts on the Deterioration of Concrete Infrastructure - Synthesis Report. 
Published by CSIRO, Canberra. ISBN978 643 10364 1 

World Business Council on Sustainable Development (WBCSD). (2010). Cement 

Sustainability Initiative. Retrieved from 

http://www. wbcsdcement.org/index.php ?option=com content&task=view&id=174&Item 

id=232 

Wang, J. C. and Yan, P. Y. (2006). Influence of initial casting temperature and dosage of 
fly ash on hydration heat evolution of concrete under adiabatic conditions. Journal of 
Thermal Analysis and Calorimetry, 85(3), 755-760. 

World Business Council on Sustainable Development (WBCSD). (2005). C0 2 
accounting and reporting standard for the cement industry. Retrieved from 
http://www.wbcsdcement.org/index.php/publications 

World Business Council on Sustainable Development (WBCSD). (2005). Guidelines for 
the selection and use of fuels and raw materials in the cement manufacturing process. 
Retrieved from http://www.wbcsdcement.org/pdf/tf2 guidelines.pdf 

Zhang, F., Shen, D., Zhou, J., Li, Z. (2011). Effect of thermal environment at early age on 
hydration phases composition and strength development of concrete containing fly ash. 
Advanced Materials Research, 168-170, 582-588 

Zhang, Y. M., Sun, W., & Yan, H. D. (2000). Hydration of high-volume fly ash cement 
pastes. Cement and Concrete Composites, 22, 445-452. 

Zhou, N. and McNeil, M. A. (2009). Assessment of Historic Trend in Mobility and 
Energy Use in India Transportation Sector Using Bottom-up Approach. Lawrence 
Berkeley National Laboratory (LBNL-2415E). Retrieved from 
http://china.lbl.gov/sites/china.lbl.gov/files/LBNL-2415E.pdf 



246 



Appendix A 



Table A. la Pervious Concrete Literature 



Pervious Research 



Date 



Title 



Author 



Objectives 



Field and Laboratory 
2009 Evaluation Of Pervious 
Concrete Pavements 



2010 



2009 



2009 



2008 



2010 



Delatte, N., Mrkajic, 
A., Miller, D.I. 



Removal of Heavy Calkins, J., Kney, A., 

Metals using Pervious Suleiman, M. T., 

Concrete Material Weidner, A. 

Pervious Concrete 

Pavement Integrated Henderson, V., Tighe, 

Laboratory and Field S. L., Norris, J. 

Study 

Evaluation of Pervious 

Concrete Workability Kevern, J. T., Schaefer, 

Using Gyratory V. R., Wang, K., 

Compaction 



Study compared cores from field placement of Portland Cement Pervious Concrete with lab 

made cores. Properties included void ratio, hydraulic conductivity, compressive strength, 

compressive strength and tensile strength. The field measurements included drainage and 

indirect-transmission ultrasonic pulse velocity 

Pervious concrete mixtures were modified to improve strength and filtration properties 

simultaneously. The study involved adding fiber to the concrete mixture, which increased 

strength and increased concrete's ability to remove copper. 

Studied the performance of pervious concrete in various climatic regions of Canada. The 

applications were for parking lots, shoulders, and lanes. Also discussed was rehabilitation/ 

maintenance methods, analysis of strain caused by environmental conditions within the 

pervious concrete layer, and filtration abilities of pervious concrete. 

The Superpave gyratory compactor SGC was modified to develop a test method to 
characterize the workability of pervious concrete. 



A novel approach to 

characterize entrained Kevern, J. T., Schaefer, natural and synthetic air entraining agents were used. The RapidAir system is an automatic 

air content in pervious V. R., Wang, K., device that determines air voids according to ASTM C457 

concrete 



Maintenance and 
Repair Options for 
Pervious Concrete 



Kevern, J. 



This paper discussed common causes and identification of common and not so common 
pavement distresses. Methods were used to assess surface condition and permeability. 
Cleaning and surface repair were also explored . 



ho 



Table A. lb Pervious Concrete Literature cont. 



Pervious Research 



Date 



Title 



Author 



Objectives 



2010 



2009 



2010 



2008 



2010 



Potential for clay 
clogging of pervious 
concrete under 
extreme conditions 

Laboratory evaluation 
of permeability and 
strength of polymer- 
modified pervious 
concrete 

Effect of rejuvenation 
methods on the 
infiltration rates of 
pervious concrete 
pavements 

Solid material 
retention and nutrient 
reduction properties 
of pervious concrete 
mixtures 

Effective curve 
numberand 
hydrologic design of 
pervious concrete 
storm-water systems 



Haselbach, L. M. 



Huang, B., Wu, H., Shu, 
X., Burdette, E. G. 



Chopra, M., Kakuturu, 
S., Ballock, C, Spence, 
J., Wanielista, M. 



Luck, J. D., Workman, 
S. R., Coyne, M.S., 
Higgins, S. F. 



Schwartz, S. S. 



Research focused on mimicking a series of catastrophic clogging cycles with clay runoff 



This study focused on the balance between permeability and strength properties of polymer- 
modified pervious concrete (PMPC). In addition to latex, natural sand and fiber were included 
to enhance the strength properties of pervious concrete. 

The study included field and laboratory investigations to evaluate the infiltration capacities of 
the pervious concrete cores and the underlying soils and the usefulness of rejuvenation 
methods in restoring their hydraulic performance. A new field test device, called the 
embedded ring infiltrometer, was also tested 

Laboratory tests were conducted on replicated samples of pervious 
concrete made from two aggregate sources (river gravel and limestone) with two size 
fractions from each aggregate. Water was filtered through composted beef cattle manure 
and bedding (compost) that was placed on top of the pervious concrete specimens. 

The paper presented a procedure for consistent design and hydrologic evaluation of pervious 
concrete storm-water management systems. Design parameters of sub base thickness and the 
size and elevation of drains were identified to satisfy basic operational criteria based on 
freeze-thaw risk and the timely drawdown of sub-base storage. 



to 

oc 



Table A.lc Pervious Concrete Literature cont. 



Pervious Research 



Date 



Title 



Author 



Objectives 



2009 



2010 



2006 



2009 



2010 



2009 



Use of soft computing 
applications to model 
pervious concrete 
pavement condition 
in cold climates 
Pervious Concrete 
Testing Methods 
Vertical porosity 
distributions in 
pervious concrete 
pavement 

Temperature Behavior 
of Pervious Concrete 
Systems 

Surface temperature 
and heat exchange 
differences between 
pervious concrete, 
and traditional 
concrete and asphalt 
pavements 



Golroo, A., Tighe, S. 



Haselbach, L. 



Haselbach, L. M., 
Freeman, R. M. 



Kevern, J. T., Schaefer, 
V. R., Wang, K., 



Flower, W., Burian, S. 
J., Pomeroy, C. A., 
Pardyjak, E. R. 



This paper proposed three soft computing methods: fuzzy sets, the Latin Hypercube 
Simulation technique, and the Markov Chain process to develop a probabilistic versus 
deterministic performance curve. 

This paper reviewed some testing methods under development for pervious concrete and 
summarize research methodologies. 

The study showed the increase in porosity when slabs approximately 15 cm (6 in.) in height 
were placed with an approximately 10% surface compaction technique. A series of vertical 
porosity distribution equations were developed to effectively include the percent 
compaction and average cored porosities. 

Temperature sensors were installed through the profile of a pervious concrete pavement and 
traditional concrete pavement and into the underlying soil to monitor the temperature of 
both type of pavements. 

Surface and internal temperatures were monitored at a pervious concrete site, an adjacent 
traditional concrete site, and a traditional asphalt pavement site. The results of the surface 
and internal temperature monitoring of the pervious concrete were used to calibrate and 
validate a numerical heat flux model. 



The effect of curing 
regime on pervious 
concrete abrasion 
resistance 



Kevern, J. T., Schaefer, 
V. R., Wang, K., 



This paper presents results of combinations of four different pervious concrete mixtures 
cured using six common curing methods. 






Table A. Id Pervious Concrete Literature cont. 



Pervious Research 



Date 



Title 



Author 



Objectives 



2008 



2009 



2008 



2008 



2010 



Evaluation of an 
infiltration best 
management practice 
utilizing pervious 
concrete 

Experimental study of 
pervious concrete on 
parking lot 



Kwiatkowski, M., 
Welker, A. L, Traver, 
R. G., Vanacore, M., 
Ladd, T. 



The study measured infiltration and soil and groundwater contamination from the 
performance of a pervious concrete infiltration basin installed in a northeastern 
climate 



Lee, M. G., Chiu, C. T., A suitable mix design was tested for use in a pervious concrete lot placement. The study 
Kan, Y. C, Yen, T. evaluated strength and field permeability. 



Temperature 

response in a pervious 

concrete system Kevern, J. T., Schaefer, 

designed for V. R. 

stormwater treatment 



The paper presents data obtained from a fully instrumented pervious concrete parking lot. 
Temperature sensors monitored the freeze-thaw behavior of the system for both pervious 
sections and a standard concrete control. 



A novel approach to 

characterize entrained Kevern, J.T. ; Wang, K.; The study used a device called the RapidAirSystem to determine the entrained air voids in 

air content in pervious Schaefer, V.R. pervious concrete. 

concrete 

Effect of coarse 

aggregate on tne Kevern, J.T. ; Wang, K.; The paper showed how 17 different coarse aggregates were tested for freeze thaw durability 

Treeze-tnaw durability s cnae f er; y.R. properties and impact of angularity when using the aggregates in a pervious concrete mix. 

of pervious concrete 



to 

o 



Table A.le Pervious Concrete Literature cont. 



Pervious Research 



Date 



Title 



Author 



Objectives 



Experimental study on 
properties of pervious 
2003 concrete pavement Yang, J.; Jiang, G. 
materials 



2010 



2010 



2007 



2010 



Used smaller sized aggregate, sil ica fume (SF), and superplasticizerfSP) in the pervious 
concrete to enhance the strength of pervious concrete to reach a maximum compressive 
strength of about 50 MPa (7251.9 psi) 



Parasivamurthy, P.; 
Jawali,V.;Kirankumar, 



Improvingground 

water recharge using 

pervious cement 

concrete made of 

aggregates recycled B .V.; Patil, M. B. 

from crushed concrete 

wastes 

Compressive behavior 
of pervious concretes 
and a quantification of 



Tested strength and permeability of pervious concrete mixes incorporating the use of 
recycled aggregate 



the influence of 
random pore structure 
features 



Deo, O.; Neithalath, N. 



Tested and modeled pervious concrete with different grading of coarse aggregate such as#S, 
#4, 3/8" rock 



Aggregate effects on 

pervious pom ana Crouch, L.K. ; Pitt, J.; The effect of aggregate gradation, amount, and size on the modulus of elasticity of pervious 

cement concrete static , 



modulus of elasticity 



Hewitt, R. 



concrete mixtures were compared 



The study recorded temperatures of incoming stormwater runoff, porous pavement surface 
temperature t ^ e j nf j 1 1 rate cl water that collected atthe bottom of the infiltration bed in order to determine 

mitigation Deneatn Barbis, J.; Wei ker, A. L whether porous pavements and the infiltration bed were effective i n reducing high 
porous pavements temperature runoff. 






Table A.2 A Comparison of Literature Regarding Hot Weather Concreting or Thermal Properties of Fly Ash and High 
Volume Fly Ash Concrete. 











Lab or Field 








Percentage of fly 


Relative 






Author 


Published Year 




Curing 


Additional Testing Variables 






ash used 


Humidity 


Temperature 














Phase 1 - water content kept 








65% or water 


20°C (68°F), 40°C 




Ravina, D. 


1981 


0%, 20%, 30% 






constant, Phase II - slump 








cured 


(104°F) 


kept constant 












Maintaining a rise in 












internal curing temperature 










25°C (77°F) to 


between 15°C(27°F)-30°C 


Merita, P. K. 


2002 


57% 


80%- 100% 


30°C(86°F) 


(54°F) for massive concrete 
structural members in order 
to prevent thermal cracking 








Unknown 




Measure the variation of 








(testing occurred 














temperature overtime, 








in humidity 




strength with different 


Senthil, S. & 




0% and Blended 


controlled 




curing and heat dissipation 


Santhakumar, A. 


2005 


cement 


chamber), water 


Unknown 


methods, and temperature 


R. 




(unknown %) 


curing and a 

water curing 

compound were 

also used 




at different depths of a 

sample with aim height 

3.28 ft 








Some cured in a 






Bentz, D. P., 






lime solution 




Measure specific heat 


Peltz, M. A., 




0%, 15%, 30%, 


and others in 




capacity, thermal 


Duran-Herrera 


2010 


45%, 60%, 75% 


sealed plastic 


25°C(77°F) 


conductivity (transient 


A., Valdez, P., 






conditions in 




plane method) of mortars 


Juarez, C. A. 






40% humidity 




and concretes 



to 
to 



Appendix B 



Table B.l Exam] 


pie of Fuel Emission Factors from Various Sources 


Fuel/Electricity 


India and IPCC 
(CCAP, TERI, 
2006 based on 
IPCC, 1996 and 
MoEF, 2004) 


EPA (2008) 


Cement 
Sustainability 

Initiative 
(WBCSD, 2010) 


India Specific 
(MoEF, 2010) 


US Department 
of Energy, 

Energy 

Information 

Administration 

(EIA, 2007) 


Clean 

Development 

Mechanism 

Electricity 

Average India 

(CDM, Bhat, 

2006) 


EF(kgC02/GJ) 


EF (kg C02/GJ) 


EF(kgC02/GJ) 


EF(kgC02/GJ) 


EF(kgC02/GJ) 


EF (kg C02/GJ) 


Crude oil 


72.60 












Aviation turbine fuel (ATF) 


70.79 






71.5 






Diesel 


73.33 






74.1 






Gasoline 


68.61 






69.3 






Fuel oil/residual fuel oil 


76.59 


74.69 




77.4 






Kerosene 


71.15 


68.54 




71.9 






Natural gas 


55.82 


50.29 










Naphtha* 


72.60 












Gas/diesel oil 


73.33 












LPG 


62.44 


59.86 




47.3 






Lignite 


93.10 


91.40 




106.15 






Non coking coal domestic 


78.65 






95.81 






Non coking coal imported 


88.38 












Coking coal prime domestic** 


84.33 


88.58 




93.61 






Coking coal inferior domestic 


84.33 












Imported coking coal 


87.03 












Antrhacite coal 




98.21 










Sub-bitumimous coal 




92.02 










Unspecified (industrial 
coking) 




88.83 










Unspecified (industrial other 
coking) 




89.08 










Unspecified (electric utility) 




89.52 










Unspecified 
(residential/com me rical) 




90.36 










Coke 




107.74 










Distillate Fuel Oil (#1,2,4) 




69.33 










Residential Fuel Oil (#5, 6) 




74.69 










Petroleum Coke 




96.79 


92.8 








Ethane 




56.47 










Propane 




59.78 










Isobutane 




61.68 










n-Butane 




61.58 










Waste Tires 




106.95 


85 








Waste oil 






74 








Plastics 






75 








Solvents 






74 








Impregnated saw dust 






75 








Mixed industrial waste 






83 








Other fossil based wates 






80 








Dried sewage sludge 






110 








wood, non impregnated saw 
dust 






110 








paper, carton 






110 








animal meal 






89 








animal bone meal 






89 








animal fat 






89 








agricultural, organic, diaper 
waste, charcoal 






110 








Other biomass 






110 








Compressed Natural Gas 
(CNG) 








56.1 






Lubricants 








73.3 






Electricity Grid 










275 


233.3 



253 



m 










SiHh^*» 






CEMENT 








FORM -A {SEE RULE 2} 








form of Disclosure of particulars whr 1 respect to Conservation of Energy) 








Sr.No. Particulars 


2010-11 


£008-2010 






12 months 


13 months 




A POWER & FUEL CONSUMPTION 








1 Electricity 








a) Purchased 








Unit(Kwn) -Lacs 


1061.77 


1416.95 




Total Amount (? in Lacs) 


6766.16 


9231.35 




Rate f Unit (?) 


6.25 


6.52 




b) Own Generation 








i Through Diesel gereratcr 


1.67 


446.35 




ii Fuel Cost ..'Electricity Duty 


16.04 


2142.03 




iii Through Steam Turbine .'Generator 


Ml 


Nil 




2 Cool & Other Fuel Used 33 Kiln/Fuel 








quantity { in Million K.CaQ 


037363 


1604727 




Total Cost (? in Lacs) 


9706.70 


12437.10 




Average rate (? in Million K.CaJ) 


1035.85 


775.03 




3 H.S.D. i L.D.O 








Quantity in (KLtrs) 


184.70 


200.93 




Total Cost of (T in lacs) 


75.63 


73.07 




Average rate (? in K.Lrrs) 


40926.35 


36356.31 




4 Other .Internal Generation 


Nl 


Nil 




D CONSUMPTION PER UNIT OF PRODUCTION 








Electricity (Kwh/T of Cements* 


86.21 


BS.62 




Diesel (Ltr..T. of Clinker) 


0.16 


D.10 




Coal .'Lignite (K.Cal /Kg. ol Clinker) 


sn 


738 




Others 








T Net of nor productivity of Power 







Figure B.l Typical Cement Company Data on Fuel and Electricity Consumption 
from Annual Reports 



254 



Appendix C 



Afe^-. 1 ABORATORY TEST REPORT 

Sftetttmsrttot&tmttmrtTwr CLIENT; Beslway Concnelfl WusT<sxt PROJECT NO • 202(1 OH 
S«5Na™ P Stn*! SOURCE: Snghlcn. REPORT DATE: March 5, 2M8 
EX.iw«, CO 802M SAMPLED BY; Client 
Ja3 9/5 9959.F« 30*975 9909 PROJECT: (Asceilaoeous 


MAIfcRIAL 
DESCRIPTION 


AS I'M C 33 Siie No btm coarse Aggregate 


pa re 

SAMPLED 


January 28, 2Q08 


SAMPLE 
LOCATION 


BlunUpli 


Aggregate Pineal Property and QijaMy Fa-sln [ASTM C 33 S|»o(lcH»on») 


ASTM C lir* C 136, AASHTO T 1 1 A r 27 


ASTM C 127, AASHTO T 05 Buk Spngiflr. 

Gravity - 2.S9. Sulk Specific Gravity {S5P) - 

2 61, Apparent Specific Gminly = 2.6*1. 

Absorption = 0-B% 


ASTM C BB, AASHTO T 104, Sodium S.prL'.lt: SouiiUiiais: 
5 Cycles 


5IEVESIZfc 


■•!'■■■: r:i; 


Si--.' ,-. , ii,i 


No .67 
Sr«rtflC«tlO»j 


SIEVE 
StZE 


GRADING 

OF 
■ RVSMM 

SAMPLE 


WEIGHT 
BEFORE 

TEST a 


Iv-ikl HI 
PASSING 

-.- hi-; 
TCST 


WEWHTEE 
PERCENI 

< ■■■::■- 


r 








ASTM C 131. AASHTO T 9G, I A Ahrasion 
Or3dif>3 B, Lasts - 4Q">ii 
SPKiricalicW 45% Ma* 


i- „/ 


100 


too 




r 


100 


65- 100 


1 00 


M,.-n r 


9 




1 s 


0.1 


3T4- 


92 




90- 100 


ASTM C 1<12, AASHTO T 112. Clay Lumpi* 
Friable Particles 

COARSE AGO - 0.2%, SpotlhcalKjn 3 0*>i MAX 


riotf*" 


9 14 7 


1«" 


60 


£5 -60 




:««• ip 1J2 - 


M 


sr?fl 


OB 


0.4 


sap 


40 




20 56 




WftMM* 


3J14 


I .- 


7 


0-111 


Q. 1U 


ASTM C 123. AASHTO T 113. Ughlweiylit 
Particles in AggrigQata 


:mt io r* 4 


3S 


SOQ 7 0.3 


0.1 


UB 


-■■ 


0- S 


i- 


Ttmi 


100 


C0*UW6«16 IUI1HW1 


1 


* 16 


2 








LIQUID ."VII . 
' 


UQMMrVOMT 

f>jutna.c» 


EPFC 


specification 


12 M»s. 


*»3Q 


2 






ASTM C 2S. AASHTO T 10. 
Fkjlfc Density and Vrxrfa in Aggregate 

Redding Method, Bulk Densily ■ 103 pot 
Voids in AyoreQaile ■ afiti 


§80 


1 






3108 7 


ZnClj/20 


0.0% 


0!/!: Mm- 


tt 100 


1 






J T '1H / 


2nBf,tt-4 


0,0% 


3 0% Max. 


»20O 


ob 


0-1.5 


1.5 




COMMfcNIS ' """ 



Figure C.l Coarse Aggregate Sieve and Other Laboratory Analyses 



to 







845 NmDfO Slfeel 

Lleoua. CO 80204 

3D3 97 , 5.B95!i fa* 303 97S S96S 


LABORATORY TEST REPORT 

CUENT: Beslwsy Concrete WesTest PROJECT NO.: 2024438 
SOURCE: Brighton REPORT DATE: March 5, 2006 
SAMPLED BY: Client 
PROJECT: Miscellaneous 




MATERIAL 
DESCRIPTION 


ASTM C 33 Fine Aggregate 


D'lTI: 

SAMPLED 


JfMWry 28,2001 


LOCATION 


SloefcpilB 


Aflrjroqate Pdy^ral Property arxl QuaMy Trsrts (ASTM C 33. AASHTO M 8 Spadficalioro; 


ASTM C 1 17 


A C 138. AASHTO T 11 « T 27 


ASTM C 128, AASHTO T 84. Bulk Specific 

Gwlly = 2.61 . Bulk Specific Gravity (SSO) = 

2.03. Apparent Specific Gnwily ■ 1 66 

Absorption ■ D.7% 


AST".4 r . ri." WtjTO 1 If hi " ■ ----- .■ i. -,ir:r -■ ,,. ..I--,:-.,. 
Spyules 


SIEVE 


GRADING 

OF 
ORIGINAL 
SAMPLE 


WEIGHT 
BEFORE 
TEST.g 


PERCENT 

PASSING 

AFTER 

nun 


WEIGH lhlj 

PERCENT 

LOSS 


SIEVE 5<ZE 


% Passing 


ASTM C 

33 Sp«: 


AASKIO 
M 6 Spec. 


V 








ASTM D 2419, AASHTO T 176. 

Stuui Fquivalani Value * so 

SpccificaUin: HO Mm (CDOT) 


3/4" 








fc%ii«»*l00 


4 








1/2* 








e 50 ia( u» 


10 








3/8" 




100 


100 


ASTMC 142, AASHTO T 112. Clsy Lumps & 
I INC A.GG = 0,7%, SpecfhCHltoo 3.0H Ma*, 


#30 to II 3U 


24 


iDD.D 


4.0 


10 


# -i 


loo 


»6- KXI 


95 100 


s WLuwau 


31 


MM 


u 


0,4 


•a 


99 


so. too 


M- 100 


SB IHB IE 


2a 


100 


2 


06 


#16 


71 


SO 6'j 


50-^6 


ASTM C 123. AASHTO T 1 13, Lk-hlw«lgr>l 
Particles In Aggregate 


M to* 4 


i 




2.0 


J LI 


U 30 


38 


25 00 


25-60 


3ffl- tofl< 










MSO 


14 


5-30 


5-30 


SAMPLE 
WIT 


inuiDTrPEi 

SPECIFIC 


UtiHlWHOMT 

f*khcu=s 


SPF.C, 


IOIAI 


100 


TlNEAC-G TOTAL 100% 


1 


#100 


4 


Ll 10 


C ifj 


SPE'CIFI 


CATION 




10 Max 


# 2*,: 


14 


0-3 


i) :' 


?1!l 1 


ZnCV2 


mi 


0.6% Mat 


ASTM C 40, AA.SHTO T 21, Oyanic Impuritee: 

Less Iran Organic Piste No. 1 

SpcttafciiUon: Organic Plain No. 3 or Less 


rincnass 

Modulus 


274 


2.3-3.1 




210.1 


2n6fj.'2 4 


0.0% 


3 0% Max. 


COMMENTS: " " " — 

1 







Figure C.2 Fine Aggregate Sieve and Other Laboratory Analyses 



to 

ON 




WATER COLLECTION TANK OR OTHER TYPE OF DRAIN ■ 



150 MM THICK PERVIOUS CONCRETE 

i — 150 MM THICK BASE COURSE 



0.45 



'•*• '■'■■' '" ' ■' '■ ■ "■ ■■' ' ■'■ ■••'--' '• -■ -*'■ '• '' ■ •' ■•' - ■"■ • ■■■-'-'•-'-■ • . -"-■. :■■ • •-.-■■. -.. ■■' .-..■-■..■ -.-..--••■■ . . ■ 



0.20 



TH 






0.30 






(c) 



NOTE: 

1 . All dimensions are in 
meters unless noted 
otherwise 

2. Base C ourse is good 
quality black metal with no 
fines 

3. Perforated pipeshaN have 
a 1% [min)slope 

4. Perforated pipe shall be 
wrapped with geotextile fiber 
to prevent fine aggregate 
[sand) from dogging slots 
otherwise use sam e base 
course material to fill trench 

5. Pervious concrete shall be 
surrounded by a brick wall 
that is level with the pervious 
concrete. The brick is a 
protective barrier from fines 
clogging the pervious 
concrete. 

6. A2 to 3 cm layer ofsand 
shall be placed between the 
geotextile fiber and base 
course to provide cushion 



to 



^-00.10 
150 MM THICK SAND 
2 MM THICK IMPERMEABLE MEMBRANE 
Figure C.3 Autocad Drawing of the (a) Layout and (b) Profile of the Pervious Concrete System (c) Close-Up of Profile 



Appendix D 




(a) (b) 

Figure D.l Vanakbori (a) Compressive Strength Testing (b) Fracture Paths 




(c) (d) 

Figure D.2 Gandhinagar (a) Compressive Strength Testing (b) Fracture Paths 





3T V 



(a) (b) 

Figure D.3 7-Day Compressive Strength Testing Fracture Paths (a) Cylinders and 
(b) Cubes 



258 




(a) (b) 

Figure D.4 28-Day Compressive Strength Testing Fracture Paths (a) Cylinders and 
(b) Cubes 




(a) (b) 

Figure D.5 56-Day Compressive Strength Testing Fracture Paths (a) Cylinders and 
(b) Cubes 




Figure D.6 OPC65W 56-Days Voids/ But Paste is Smoother 



259 




rii 




(a) (b) 

Figure D.7 Fracture Pattern for OPC Water Cured Samples (a) 3-Days OPC55W, 
(b) 90-Days OPC55W 




(a) (b) (c) 

Figure D.8 Fracture Patterns and Texture for OPC55H at 90 Days 




Figure D.9 50FA55W at 1 Day 



260 




(a) (b) 

Figure D. 10 50FA65W (a) at 28 Days (b) Side Fracturing Occurring up Until 56 
Days of Testing 




Figure D.ll 50FA55H Powdery at 56-days 




Figure D.12 Early Versus Later Age Breaking for Heat Cured Fly Ash Samples 
(a) 7-Day 50FA55H (b) 90-Day 50FA55H 



261 




(a) (b) 

Figure D.13 Early Versus Later Age Breaking for Heat Cured OPC Samples 
(a) 3-Day OPC65H (b) 90-Day OPC65H 

Heated Aggregate 



1 1 










lb i 


4^HK 


iffir 


^H 


I 

Li ' 








i * 








^^ ,4 Vj 








^Fj 


^ 













11 






, 



(a) (b) 

Figure D.14 Texture of Water Cured OPC (OPC55W) Samples (a) Fracture Pattern 
(b) Close-Up of Texture Pattern 




(a) (b) 

Figure D.15 (a) OPC55H_HA Porous (b) OPC55H_HA Characteristic of a 
Stalagmite at 1-Day 



262 




Figure D.16 Texture of Fly Ash Water Cured (50FA65W) Samples (a) 90-Days 
Fracture and (b) Close up of Texture 




(a) (b) 

Figure D.17 Texture of Fly Ash Heat Cured (50FA65H) Samples (a) 90-Days 
Fracture and (b) Close up of Powdery Texture 



263 



Regression Analysis: X versus A, B, C, D, T 



The regression equation is 

X = 6563 - 1777 A - 379 B + 468 ( 

Standard Error of 
Predictor Variable 

= S/(Spkpk) 



Predictor 

Constant 

A 

B 

Z 

D 

r 



157 D + 6.48 T 

T-Test statistic for 
testing Hoip^O 



P- value for testing 
H o :p 1= 



Coef 



6562 

-1777. 

-378 

467. 
-157. 

6.4! 




|ik|ik 



Ep k 2 -(2:Pk) 2 /n 



T = Pk/SEPk 



Standard deviation of 
residuals 



= 540.204 R-Sq 

Analysis of Variance 
Degrees of freedom for 
confidence intervals and 
significance tests 



Coefficient of 
determination, 



R z = SSR/SST*10(> 




Adjusted value, the proportion of 
the variance of response explained 
by predictors 



.2% 



R-Sq(adj) 



'5.6% 



Sum of Squares 



Mean 
Squares = 
SS/DF 



Source \ DF 
Regression M 5 
Residual Error 42 
Total 47 



SS 
3840803. 
?25645< 



Source 

A 
B 
Z 
D 

r 



DF 
1 

1 
1 
1 
1 



Seq SS 

37899264 

1721671 

2627664 

296154 
1296050 




MS 
768161 
291820 




F-Statistic = 
MSR/MSE, 
testing that all 
coeff. are zero 



p-value to 
determine 
significance, 
if < a=0.05 then 
significant 



SSR = £(X actual - X bar ) 

SSE = ^(Xpfedjct,,,! — X bar ) 

SST = £(X actua i — Xp red [ cted ) 



Sequential Sum of Squares 



Unusual Observations 

Obs A X 
28 0.00 5100.0 620 



/ 



Fit 


SE Fit 


Residual 


St Resid 


8.3 


197.3 


-1108.3 


-2.20R 



Standardized residuals 
greater than 2 or less than 
-2 identify an outlier and 
are calculated by dividing 
the residual by the standard 
deviation. (Refer to p. 720 
in Hayter (1996) for 
standard deviation 
procedures) 



R denotes an observation with a large standardized residual. 



Figure D.18 I s Regression Analysis for X 



264 



Regression Analysis: Y versus A, B, C, D, T 

The regression equation is 

Y = 3084 - 467 A + 1366 B - 377 C + 86 D - 5.77 T 



Predictor 


Coef 


SE Coef 




T 




P 


Constant 


3084.1 


376.7 


8 


19 





000 


A 


-467.2 


235.5 


-1 


98 





054 


B 


1366.3 


235.5 


5 


80 





000 


C 


-376.5 


235.5 


-1 


60 





117 


D 


86.1 


235.5 





37 





717 


r 


-5.767 


4.645 


-1 


24 





221 



S = 815.909 R-Sq = 49.9% R-Sq(adj) = 43.9° 
Analysis of Variance 



Source 




DF SS 


MS 


F 


P 


Regression 


5 27837042 


5567408 


8.36 


0.000 


Residua 


1 Error 42 27959682 


665707 






Total 




47 55796723 








Source 


DF 


Seq SS 








A 


1 


2619245 








B 


1 


22401629 








C 


1 


1701303 








D 


1 


88909 








r 


1 


1025956 









Unusual Observations 

Cbs A Y Fit SE Fit Residual St Resid 
19 1.00 3617 2079 298 1538 2.02R 

29 0.00 6069 4214 263 1855 2.40R 

30 0.00 6517 4017 302 2499 3.30R 

R denotes an observation with a large standardized residual. 



Figure D.19 I s Regression Analysis for Y 



265 



Regression Analysis: Z versus A, B, C, D, T 



The regression equation is 



Z = 0.00569 


+ 0.0181 A 


- 0.0384 B + 


0.00070 


C 


- 0. 


00649 D - 


- 0.000095 


Predictor 


Coef 


SE Coef 


T 




p 






Constant 


0.005686 


0.005790 


0.98 





332 






A 


0.018115 


0.003620 


5.00 





000 






B 


-0.038448 


0.003620 


-10.62 





000 






C 


0.000698 


0.003620 


0.19 





848 






D 


-0.006490 


0.003620 


-1.79 





080 






r 


-0.00009506 


0.00007139 


-1.33 





190 







S = 0.0125394 R-Sq = 77.3% R-Sq(adj) = 74.6% 



Analysis of Variance 



Source 




DF 


SS 


MS 


F 


P 


Regression 


5 


0224666 


0.0044933 


28.58 


0.000 


Residua 


1 Error 42 


0066039 


0.0001572 






Total 




47 


0290705 








Source 


DF 


Seq SS 










A 


1 


0.0039377 










B 


1 


0.0177389 










C 


1 


0.0000058 










D 


1 


0.0005054 










r 


1 


0.0002788 











Unusual Observations 

Obs A Z Fit SE Fit Residual St Resid 

22 1.00 0.01850 -0.01661 0.00458 0.03511 3.01R 

23 1.00 0.00875 -0.01927 0.00405 0.02802 2.36R 

R denotes an observation with a large standardized residual. 



Figure D.20 I s Regression Analysis for Z 



266 



Regression Analysis: X versus A, B, C, D, T, T*B 



The regression equation is 

X = 6042 - 1777 A + 663 B + 468 C 



Predictor 


Coef 


SE Coef 


T 




p 


Constant 


6041.9 


277.1 


21.80 





000 


A 




-1777.2 


140. 9 


-12.61 





000 


B 




663.0 


351.8 


1.88 





067 


C 




467 .9 


140. 9 


3.32 





002 


D 




-157.1 


140. 9 


-1.11 





271 


r 




15.462 


3.930 


3.93 





000 


r*B 




-17 .961 


5.558 


-3.23 





002 


s = 


488.115 


R-Sq 


= 82.6% 


R-Sq(adj) 


= 8 



157 D + 15.5 T 



18.0 T* 



80.0% 



Analysis of Variance 

Source DF SS MS F P 

Regression 6 46328739 7721456 32.41 0.000 

3768524 238257 



Residual Error 41 



Total 



Source DF 

A 1 

B 1 

C 1 

D 1 

r l 

r*B l 



47 56097263 



Seq SS 

37899264 

1721671 

2627664 

296154 
1296050 
2487936 



Unusual Observations 

Cbs A X Fit SE Fit Residual St Resid 
28 0.00 5100.0 6477.8 196.8 -1377.8 -3.08R 

R denotes an observation with a large standardized residual. 



HOT 



Figure D.21 2 Regression Analysis for X 



267 



Regression Analysis: Y versus A, B, C, D, T, T*B 



The regression equation is 

Y = 4032 - 467 A - 529 B - 377 C + 86 D 



Prec 


ictor 


Coef 


SE Coef 


T 




p 


Constant 


4031.5 


393.8 


10.24 


0.000 


A 




-467 .2 


200.2 


-2.33 





025 


B 




-528.6 


500.0 


-1.06 





297 


C 




-376.5 


200.2 


-1.88 





067 


D 




86.1 


200.2 


0.43 





670 


r 




-22.102 


5.585 


-3.96 





000 


r*B 




32.670 


7.899 


4.14 





000 


s = 


693.672 


R-Sq 


= 64.6% 


R-Sq(adj 





Analysis of Variance 



22.1 T + 32.7 T* 



59.5% 



Source 
Regression 



DF 
6 



SS 



MS 



36068334 6011389 12.49 0.000 



Residual Error 41 19728390 



481180 



Total 



Source DF 

A 1 

B 1 

C 1 

D 1 

r l 

r*B l 



47 55796723 



Seq SS 
2619245 
22401629 
1701303 
88909 
1025956 
8231292 



Unusual Observations 

Cbs A Y Fit SE Fit Residual St Resid 

29 0.00 6069 4181 224 1888 2.88R 

30 0.00 6517 4540 286 1976 3.13R 

R denotes an observation with a large standardized residual. 



,nd 



Figure D.22 2 Regression Analysis for Y 



268 



Regression Analysis: Z versus A, B, C, D, T, T*B 

The regression equation is 

Z = - 0.00160 + 0.0181 A - 0.0239 B + 0.00070 C - 0.00649 D + 0.000030 T 
- 0.000251 T*B 



Prec 


ictor 


Coef 


SE Coef 




T 


p 


Cons 


tant 


-0.001596 


0.006934 


-0 


23 


0.819 


A 






0.018115 


0.003526 


5 


14 


0.000 


B 






-0.023886 


0.008805 


-2 


71 


0.010 


C 






0.000698 


0.003526 





20 


0.844 


D 






-0.006490 


0.003526 


-1 


84 


0.073 


I 






0.00003047 


0.00009836 





31 


0.758 


r*B 






-0.0002511 


0.0001391 


-1 


80 


0.078 


s = 





0122153 R-Sq = 


79.0% R-Sc 


T(ac 


ii) 


= 75.9 



Analysis of Variance 



Source 




DF 


SS 


MS 


F 


P 


Regression 


6 


0229527 


0.0038254 


25.64 


0.000 


Residua 


1 Error 41 


0061178 


.0001492 






Total 




47 


0290705 








Source 


DF 


Seq SS 










A 


1 


0.0039377 










B 


1 


0.0177389 










C 


1 


0.0000058 










D 


1 


0.0005054 










r 


1 


0.0002788 










r*B 


1 


0.0004861 











Unusual Observations 

Cbs A Z Fit SE Fit Residual St Resid 

22 1.00 0.01850 -0.01285 0.00492 0.03135 2.80R 

23 1.00 0.00875 -0.01902 0.00395 0.02777 2.40R 

R denotes an observation with a large standardized residual. 



Figure D.23 2 nd Regression Analysis for Z 



269 



Regression Analysis: X versus A, B, C, D, T, T*B 



The regression equation is 

X = 6014 - 1802 A + 804 B + 452 C 



Predictor 


Coef 


SE Coef 


T 




p 


Constant 


6014.0 


196.9 


30.54 





000 


A 




-1802.3 


101.2 


-17.80 





000 


B 




804.3 


252.8 


3.18 





002 


C 




452.5 


101.2 


4.47 





000 


D 




-157.1 


103.3 


-1.52 





131 


r 




15.277 


2.824 


5.41 





000 


r*B 




-18.365 


3.994 


-4.60 





000 


s = 


554.557 


R-Sq 


= 76.9% 


R-Sq(adj) 


= 7 



Analysis of Variance 



157 D + 15.3 T 



18.4 T*B 



Source DF 

Regression 6 

Residual Error 113 

Total 119 



SS 

115704778 

34751308 

150456086 



MS 

19284130 

307534 



75.7% 



F P 
62.71 0.000 



Source 

A 
B 
C 
D 

r 

r*B 



DF 
1 

1 
1 

1 
1 
1 



Seq SS 

37443152 

2041803 

6142235 

710771 
2864197 
6502620 



Unusual Observations 



Cbs 



Fit SE Fit Residual St Resid 



17 





00 


5339 





6540 


3 


141 


9 


-1201 


3 


-2 


24R 


72 


1 


00 


6355 





5190 


5 


141 


9 


1164 


5 


2 


17R 


79 





00 


5158 





6574 


7 


145 


9 


-1416 


7 


-2 


65R 


80 





00 


5042 





6574 


7 


145 


9 


-1532 


7 


-2 


8 6R 


83 





00 


4966 





6383 


2 


149 


2 


-1417 


2 


-2 


65R 


92 





00 


5379 





7027 


2 


145 


9 


-1648 


2 


-3 


08R 



R denotes an observation with a large standardized residual. 



7?3 



Figure D.24 3 Regression Analysis for X 



270 



Regression Analysis: Y versus A 


, B, C, D, 


T, T*B 


The regression equation is 






Y = 4032 - 


467 A - 


529 B - 377 C + 


86 D - 22 


.1 T + 32.7 T*B 


Predictor 


Coef 


SE Coef T 


P 




Constant 


4031.5 


330.5 12.20 


0.000 




A 


-467 .2 


168.1 -2.78 





007 




B 


-528.6 


419.7 -1.26 





211 




C 


-376.5 


168.1 -2.24 





028 




D 


86.1 


168.1 0.51 





610 




r 


-22.102 


4.688 -4.71 





000 




r*B 


32.670 


6.630 4.93 





000 




S = 823.454 R-Sq 


= 54.4% R-Sq(s 


dj) =51. 


4% 


Analysis o 


: Variance 






Source 


DF 


SS 


MS 


F P 


Regression 


6 


72136667 12022778 17. 


73 0.000 


Residual Error 89 


60348878 678077 




Total 


95 


132485546 






Source DF 


Seq SS 






A 1 


5238489 






B 1 


44803259 






C 1 


3402606 






D 1 


177818 






r l 


2051912 






r*B l 


16462584 






Unusual Ob 


servations 






Cbs A 


Y 


Fit SE Fit 


Residual 


St Resid 


35 1.00 


2232.8 


3986.9 240.5 


-1754.1 


-2.23R 


44 1.00 


1167.8 


2955.2 234.7 


-1787.3 


-2.26R 


56 0.00 


2277.0 


3885.0 234.7 


-1608.0 


-2.04R 


57 0.00 


6940.2 


4180.9 188.2 


2759.4 


3.44R 


60 0.00 


7505.2 


4540.2 240.5 


2965.0 


3.7 6R 


93 1.00 


1564.9 


3337.1 188.2 


-1772.2 


-2.21R 


R denotes an observation with a lai 


ge standa 


rdized residual. 


!?:„..„„ r» i 


r ard „„„ 


„„™:~., a „„i..„:„ i 


n 


■*/ 





271 



Regression Analysis: Z versus A, B, C, D, T, T*B 

The regression equation is 

Z = - 0.00160 + 0.0181 A - 0.0239 B + 0.00070 C - 0.00649 D + 0.000030 T 
- 0.000251 T*B 



Prec 


ictor 


Coef 


SE Coef 




T 


p 


Cons 


tant 


-0.001596 


0.006934 


-0 


23 


0.819 


A 






0.018115 


0.003526 


5 


14 


0.000 


B 






-0.023886 


0.008805 


-2 


71 


0.010 


C 






0.000698 


0.003526 





20 


0.844 


D 






-0.006490 


0.003526 


-1 


84 


0.073 


I 






0.00003047 


0.00009836 





31 


0.758 


r*B 






-0.0002511 


0.0001391 


-1 


80 


0.078 


s = 





0122153 R-Sq = 


79.0% R-Sc 


T(ac 


ii) 


= 75.9 



Analysis of Variance 



Source 




DF 


SS 


MS 


F 


P 


Regression 


6 


0229527 


0.0038254 


25.64 


0.000 


Residua 


1 Error 41 


0061178 


.0001492 






Total 




47 


0290705 








Source 


DF 


Seq SS 










A 


1 


0.0039377 










B 


1 


0.0177389 










C 


1 


0.0000058 










D 


1 


0.0005054 










r 


1 


0.0002788 










r*B 


1 


0.0004861 











Unusual Observations 

Cbs A Z Fit SE Fit Residual St Resid 

22 1.00 0.01850 -0.01285 0.00492 0.03135 2.80R 

23 1.00 0.00875 -0.01902 0.00395 0.02777 2.40R 

R denotes an observation with a large standardized residual. 



Figure D.26 3 rd Regression Analysis for Z 



272